Investigators: Andrew Martin, Chris Blyth, Mejbah Bhuiyan, Peter Richmond, Tom Snelling, Yue Wu, Julie Marsh
External collaborators: Professor David Smith (PathWest Laboratory Medicine), Meredith Borland (Emergency Department, Perth Children's Hospital), Dr Steven Mascaro, Professor Mark Nicol
Pneumonia, an acute respiratory infection, is the leading cause of hospitalization and death in young children globally. In Australia, death is thankfully rare but hospital admission is common. The infection is caused by a range of bacteria and viruses. Bacteria were previously thought to be most important, but more recently, viruses are detected more often than bacteria in children with pneumonia. Using support from the Telethon-Perth Children Hospital Foundation fund (2015-2018), we enrolled 270 children with pneumonia and equal numbers of healthy children into PneumoWA, a study designed to determine which pathogens (bugs) cause pneumonia in Western Australian children. We detected multiple pathogens from a nose swab taken from both sick and healthy children; we then compared the pathogens identified in the two groups. Some viruses were found much more often in sick children than healthy children, whereas other viruses and most bacteria were present equally in both sick and healthy children. Despite a comprehensive examination of these data, we could not find a single factor that could be used to tell if a child had bacterial pneumonia or viral pneumonia.
Identifying the disease-causing pathogen is important, because it will help clinicians determine who needs antibiotics and who does not. Clinicians usually prescribe antibiotics to treat pneumonia. Unfortunately, antibiotics do not help when pneumonia is caused by viruses, meaning many of the children in PneumoWA received antibiotics which did not help and potentially caused side effects. The unnecessary use of antibiotics contributes to a global problem called antimicrobial resistance (AMR). AMR occurs when antibiotics can no longer harm or kill the pathogen causing disease. AMR results in increased suffering, longer stays in hospital, and even death from untreatable infection.
PneumoBNA seeks to develop a tool to help clinicians predict the most likely pathogen causing pneumonia in a given child. We will use data from PneumoWA to explore the complex relationships between the development of disease, appearance of symptoms, and the body’s immune process for different pathogens. The model is expected to facilitate a rapid microbiological diagnosis (likely bacterial; likely viral; likely mixed) in pneumonia cases seen in the emergency department or the GP clinic. After the initial development of the tool, we will arrange a series of consultation workshops with primary care clinicians, infectious disease experts and community/consumers to share the preliminary results and seek advice from them on how to improve the prediction tool from the clinical perspective and make the tool useful in paediatric practice.
Funding: This work is funded by the Perth Childrens Hospital Foundation.