Rapid Metagenomic Next-Generation Sequencing during an Investigation of Hospital-Acquired Human Parainfluenza Virus 3 Infections

Abstract

Metagenomic next-generation sequencing (mNGS) is increasingly used for the unbiased detection of viruses, bacteria, fungi, and eukaryotic parasites in clinical samples. Whole-genome sequencing (WGS) of clinical bacterial isolates has been shown to inform hospital infection prevention practices, but this technology has not been utilized during potential respiratory virus outbreaks. Here, we report on the use of mNGS to inform the real-time infection prevention response to a cluster of hospital-acquired human parainfluenza 3 virus (HPIV3) infections at a children's hospital. Samples from 3 patients with hospital-acquired HPIV3 identified over a 12-day period on a general medical unit and 10 temporally associated samples from patients with community-acquired HPIV3 were analyzed. Our sample-to-sequencer time was <24 h, while our sample-to-answer turnaround time was <60 h with a hands-on time of approximately 6 h. Eight (2 cases and 6 controls) of 13 samples had sufficient sequencing coverage to yield the whole genome for HPIV3, while 10 (2 cases and 8 controls) of 13 samples gave partial genomes and all 13 samples had >1 read for HPIV3. Phylogenetic clustering revealed the presence of identical HPIV3 genomic sequence in the two of the cases with hospital-acquired infection, consistent with the concern for recent transmission within the medical unit. Adequate sequence coverage was not recovered for the third case. This work demonstrates the promise of mNGS for providing rapid information for infection prevention in addition to microbial detection.

Keywords: HPIV3; hospital-acquired infections; human parainfluenza 3 virus; infection prevention; mNGS; metagenomics; next-generation sequencing; parainfluenza; rapid sequencing; viral sequencing.

FIG 1

Whole-genome phylogeny of HPIV3 sequences revealed identical sequences in putative outbreak samples. Near-complete genomes recovered from HPIV3-positive specimens were aligned by MAFFT, and phylogenetic trees were constructed using MrBayes. Specimens are denoted by sample identifier name with the two identical clinical cases highlighted in orange. The NCBI reference genome for HPIV3 (NC_0001796) was used as an outgroup and its negative-stranded genome is highlighted in green. Consensus support values are denoted on branch labels. Normalized coverage maps are shown with areas of no coverage underlined in red and nucleotide changes relative to the outbreak sequence highlighted in black. Pairwise variant counts are depicted for three of the most closely related pairs of strains sequenced.