Biomarkers of viral and bacterial infection in rhinovirus pneumonia.

Background: Rhinovirus (RV) is often detected in children hospitalized with pneumonia, but the role of RV in causing pneumonia is still unclear. Methods: White blood cell count, C-reactive protein, procalcitonin, and myxovirus resistance protein A (MxA) levels were determined from blood samples in children (n = 24) hospitalized with radiologically verified pneumonia. Respiratory viruses were identified from nasal swabs by using reverse transcription polymerase chain reaction assays. Among RV-positive children, the cycle threshold value, RV subtyping by sequence analysis, and the clearance of RV by weekly nasal swabs were determined. RV-positive children with pneumonia were compared to other virus-positive children with pneumonia, and to children (n = 13) with RV-positive upper respiratory tract infection from a separate earlier study. Results: RV was detected in 6 children and other viruses in 10 children with pneumonia (viral co-detections excluded). All RV-positive children with pneumonia had high white blood cell counts, plasma C-reactive protein or procalcitonin levels, or alveolar changes in chest radiograph strongly indicating bacterial infection. The median cycle threshold value for RV was low (23.2) indicating a high RV load, and a rapid clearance of RV was observed in all. Blood level of viral biomarker MxA was lower among RV-positive children with pneumonia (median 100 µg/L) than among other virus-positive children with pneumonia (median 495 µg/L, p = 0.034) or children with RV-positive upper respiratory tract infection (median 620 µg/L, p = 0.011). Conclusions: Our observations suggest a true viral-bacterial coinfection in RV-positive pneumonia. Low MxA levels in RV-associated pneumonia need further studies.