2626. Rhinovirus in Children Presenting to the Emergency Department: Role of Viral Load in Disease Severity and Co-Infections

Abstract

Background: Rhinovirus (RV) quantitation by reverse transcription-quantitative PCR is limited by variable amplification efficiency across genotypes. We used a precise viral quantitation method, reverse transcription-digital PCR (RT-dPCR), to characterize the role of viral load in clinical outcomes and in viral co-infections in children presenting to a tertiary hospital emergency department (ED). Methods: Children < 18 years with respiratory symptoms for ≤ 14 days were enrolled from December 1, 2016 to December 31, 2018. Participants had nasal and throat specimens obtained and multiplex PCR testing with a commercial assay (FilmArray; bioMerieux). RV positive samples were quantified using RT-dPCR. Samples with sufficient viral load were sequenced at a 543 bp fragment of the RV VP4/ VP2 region. RV species were assigned by comparison to RV sequences in GenBank using BLAST. Clinical data were collected into REDCap. T-tests were used to compare mean viral loads between groups. Results: Of 1703 children enrolled in the ED, 697 were RV/enterovirus positive by FilmArray [median age 18 months (interquartile range 9-39 months)]. Of 590 subjects with viral load available, 276 (47%) were admitted to the hospital. Among RV mono-infections (N = 434), mean viral load did not differ between subjects admitted vs. discharged from the ED (7.03 log copies/mL for both, P = 0.97). Among admitted subjects with RV mono-infection, viral load also did not differ between subjects requiring supplemental oxygen vs. not (7.01 vs. 7.10 log copies/mL, P = 0.6). Subjects with viral co-infections had lower mean RV viral loads (6.31 log copies/mL) compared with those with RV only (7.03 log copies/mL; P < 0.001) (figure). Significantly different RV viral loads were seen with co-infections with respiratory syncytial virus (RSV), metapneumovirus (MPV) and parainfluenza (PIV), but not with influenza, adenovirus or coronavirus. In 525 sequenced samples (46% RV-A, 4% RV-B, 50% RV-C), viral load did not vary between RV viral species (P = 0.09). Conclusion: Precise viral quantitation demonstrates children co-infected with RV and RSV, MPV or PIV have lower nasal viral loads than those with RV alone. Among RV mono-infections, RV viral load was not associated with admission or need for supplemental oxygen.