Background: respiratory viral infections account for a considerable proportion of pediatric emergency room visits. Illnesses range in severity from mild upper respiratory tract infections to serious lower respiratory tract infections (LRTI). The relationship between viral load and specific viruses to clinical diagnosis made by physicians in this setting is poorly understood.
Methods: we applied a real-time, quantitative polymerase chain reaction (qPCR) panel for 13 common respiratory viruses to 195 frozen, archival nasopharyngeal aspirate specimens obtained from symptomatic children ≤ 4 months of age presenting to the emergency room. Mean total viral load and number of viruses per archival nasopharyngeal aspirate specimen were compared between LRTI (n = 70) and non-LRTI (1 or more of upper respiratory tract infection, fever, or cough) (n = 125), as were yield and concordance of qPCR results to viral culture/direct fluorescence assay (DFA).
Results: children with LRTI had significantly increased total viral load and harbored more viruses than the non-LRTI group. Respiratory syncytial virus-A and -B were significantly associated with LRTI, and parainfluenza virus-1 with non-LRTI. Individual loads of parainfluenza virus-2 and human rhinovirus were increased in LRTI versus non-LRTI. Quantitative PCR yielded more viruses (including coinfections, where a "dominant virus" was typically identified) than viral culture/DFA and documented nucleic acid from pathogens not tested by culture/DFA including human rhinovirus; coronaviruses -OC43, -229E, and -NL63; and metapneumovirus.
Conclusions: in symptomatic children presenting to the emergency room, total viral load is related to clinical diagnosis; specific viruses are associated with particular clinical diagnoses, and qPCR has a higher yield than other viral diagnostic methods.