Use of Genome Sequencing to Define Institutional Influenza Outbreaks, Toronto, Ontario, Canada, 2014-15.

Derek R Macfadden,Jonathan B Gubbay,Taryn Athey,Stephen Perusini,William P Hanage,Alireza Eshaghi,Romy Olsha,Allison Mcgeer,Aimin Li

doi:10.3201/eid2403.171499

Abstract

Adequacy of the current clinical definition of institutional influenza outbreaks is unclear. We performed a retrospective genome sequencing and epidemiologic analysis of institutional influenza outbreaks that occurred during the 2014–15 influenza season in Toronto, Canada. We sequenced the 2 earliest submitted samples positive for influenza A(H3N2) from each of 38 reported institutional outbreaks in long-term care facilities. Genome sequencing showed most outbreak pairs identified by using the current clinical definition were highly related. Inclusion of surveillance samples demonstrated that outbreak sources were likely introductions from broader circulating lineages. Pairwise distance analysis using majority genome and hemagglutinin-specific genes enabled identification of thresholds for discrimination of within and between outbreak pairs; the area under the curve ranged 0.93–0.95. Routine genome sequencing for defining influenza outbreaks in long-term care facilities is unlikely to add significantly to the current clinical definition. Sequencing may prove most useful for investigating sources of outbreak introductions.

Highlights

Adequacy of the current clinical definition of institutional influenza outbreaks is unclear
Of those, >2 positive H3N2 samples were confirmed in 87 outbreaks, and samples from 38 outbreaks had adequate volumes and primer amplification to be suitable for genome sequencing, and were selected for analysis
We found no statistically significant difference in distribution of outbreak sizes between the 2 groups (p = 0.94). In this retrospective genomic study of influenza A(H3N2) outbreaks in LTCFs during the 2014–15 influenza season in Toronto, we evaluated the potential role of genome sequencing in clinically defined outbreaks with >2 available pairs of H3N2-positive respiratory specimens

Summary

Introduction

Adequacy of the current clinical definition of institutional influenza outbreaks is unclear. Routine genome sequencing for defining influenza outbreaks in long-term care facilities is unlikely to add significantly to the current clinical definition. Current definitions for influenza outbreaks in hospitals or chronic/long-term care facilities (LTCFs) are illdefined, being typically based on >2 symptomatic patients in a 48–72-hour period and >1 microbiologic sample documented as positive for influenza [1,2]. This definition does not conclusively determine whether transmission events have occurred within the institution or if a linked outbreak is emerging. We evaluate whether influenza genome sequencing could improve understanding of the utility of current influenza outbreak definitions and whether it could play a role in routine outbreak identification

Methods

Results

Conclusion