Poster Presentation Australian Society for Microbiology Annual Scientific Meeting 2018

Overdiagnosis of rotavirus infection due to vaccine virus shedding. (#382)

Sarah Tozer 1 , David Whiley 2 3 , Sophia Ye 3 4 , Julia Clark 5 , Cheryl Bletchly 3 , Stephen Lambert 6 , Keith Grimwood 7 , Graeme Nimmo 3
  1. QPID - Childrens Health Queensland, Childrens Health Queensland, Brisbane, QLD, Australia
  2. Child Heath Research Center, University of Queensland , South Brisbane, Qld, Australia
  3. Pathology Queensland, Queensland Health, Brisbane, Queensland, Australia
  4. UQCCR - Herston, University of Queensland , Brisbane, Qld, Australia
  5. Infection Management and Prevention Services, Lady Cilento Children’s Hospital, South Brisbane, Queensland, Australia
  6. Communicable Diseases Branch, Queensland Health, Brisbane, Queensland, Australia
  7. Infectious Diseases, Griffith University and Gold Coast Health, Gold Coast, Queensland, Australia

Introduction: In a previous community-based birth cohort study, we observed that shedding of the Rotavirus RV5 vaccine virus was frequent and prolonged (several weeks duration) in the stool samples of vaccinated infants. Prompted by recent increases in rotavirus identification in Queensland, we investigated whether detection of vaccine shedding may impact upon routine rotavirus diagnosis and surveillance.

Methods: Available stool samples (n = 79) testing positive by rotavirus PCR at Pathology Queensland in the period late 2016 to mid 2017 were subject to rotavirus genotyping to distinguish RV5 vaccine virus from wildtype virus. The patients comprised 37 females and 42 males, ranged in age from 2 weeks to 89 years (average 34 years); 21 were less than 1 year old, by which age all three RV5 vaccine doses are administered.


Genotyping showed that of the 79 rotavirus-PCR positives samples, 6 were RV5, 1 comprised a mix of RV5 and wildtype virus and 46 were wildtype virus. A further 26 could not be genotyped, primarily because the viral load was too low. All seven RV5 detections were from children less than 1 year old; for this age group RV5 comprised 33% (7/21) of all detections.


These preliminary data suggest that detection of vaccine virus may lead to overdiagnosis of rotavirus infection in infants, and likely accounts for some (but not all) of the recent increases in numbers. The data highlight the need for screening methods to distinguish vaccine from wildtype virus. Studies are ongoing.

Disclosure of Interest Statement:
D Whiley receives research funding from SpeeDx Pty Ltd. No pharmaceutical grants were received for the development of this study.