Science digest

Childhood infections and inflammatory diseases are amongst the most common causes for seeking medical intervention. Often there are challenges in the diagnosis and treatment of childhood infections with the conventional approach is often slow due to the time for culture. Additionally, infections may be in inaccessible sites and conventional samples may fail to identify the infection. A recent publication from a European collaboration may provide a new direction for the diagnosis of childhood febrile illnesses (1). The multi-author publication used whole blood transcriptomic microarray, linked to medical learning incorporating 12 publicly available data sets. This included 1,212 children with 18 infections or inflammatory diseases. A further 411 children were investigated using a new RNA dataset panel. Overall, the 161 transcripts classified 18 disease categories which reflected a specific pathogen and a specific disease as well as predicting broad classes of diseases such as malaria and tuberculosis. The authors concluded that a single panel of RNA transcripts can be used to identify causative organisms resulting in fever in a nonspecific laboratory finding from a single blood sample, thereby reducing diagnostic delays and improving antibiotic use.