Suboptimal performance of rotavirus testing in a vaccinated community population should prompt laboratories to review their rotavirus testing algorithms in response to changes in disease prevalence Gary N McAuliffe a, ⁎, Susan L Taylor b , Sharon Moore a , Joanne Hewitt c , Arlo Upton a , Anna S Howe d , Emma J Best e a Microbiology Department, Labtests and Northland Pathology Laboratories, Auckland, New Zealand b Microbiology Department, Middlemore Hospital, Auckland, New Zealand c Enteric, Environmental and Food Virology Laboratory, Institute of Environmental Science and Research Limited, Porirua, New Zealand d Department of General Practice and Primary Health Care, School of Population Health, University of Auckland, Auckland, New Zealand e Department of Paediatrics, Child and Youth Health, University of Auckland, Auckland, New Zealand abstract article info Article history: Received 15 April 2018 Received in revised form 1 October 2018 Accepted 1 October 2018 Available online xxxx Keywords: Rotavirus Community Diagnostic Test False positive Rotavirus vaccine has reduced disease prevalence in many countries. Consequently, we aimed to assess the reli- ability of a rotavirus immunoassay in the community population of Auckland and Northland, New Zealand. Be- tween 22 October 2015 and 31 December 2016, 2873 fecal samples were tested by enzyme immunoassay (EIA, Rotascreen II, Microgen, UK) from 2748 patients (median age 8 years, range 0–101 years). Eighty-nine (3.1%) samples were reactive; 86 samples were tested by a second method. Rotavirus was confirmed in 49/86 (57%). Positive rotavirus EIAs were more likely to be confirmed in samples from cases ≥1 year of age (positive pre- dictive value [PPV] 61%, 95% confidence interval [CI] 50–72%, P = 0.049) and in spring/summer (PPV 67%, 95% CI 55–78%, P = 0.003). Reactive rotavirus tests required confirmatory testing regardless of demographic, vaccine, or seasonal factors; a review of rotavirus testing algorithms may be necessary in other vaccinated community populations. © 2018 Elsevier Inc. All rights reserved. 1. Introduction Rotavirus vaccine has dramatically altered the epidemiology of gas- troenteritis worldwide. In countries where the vaccine has been intro- duced, 89–100% reductions in emergency department visits, 74–90% declines in hospitalizations for rotavirus gastroenteritis, and 29–50% de- clines in “all-cause” acute gastroenteritis hospitalizations have been ob- served (Kollaritsch et al., 2015). Indirect vaccine benefits have extended to unvaccinated older children and adults (Mast et al., 2015). In New Zealand (NZ), rotavirus vaccine (Rotateq™, Merk & Co.) was instituted on the National Immunisation Schedule (NIS) from July 2014 as a 3- dose schedule for all infants at 6 weeks, 3 months, and 5 months of age, and in common with other countries, national rotavirus hospitali- zations declined by 87% within 2 years of vaccine introduction (Kelly et al., 2017). As a Southern Hemisphere nation, in NZ, prior to vaccine introduction, annual rotavirus peak activity was June–October (win- ter–spring); cases uncommonly occurred in the summer months (Grimwood et al., 2006). Following these changes in disease prevalence, the reliability of di- agnostic testing for rotavirus has been questioned, with reports that be- tween 20 and 42% of rotavirus antigen tests cannot be confirmed by molecular methods (McAuliffe et al., 2018; Roczo-Farkas et al., 2015; Ye et al., 2013). These findings suggest that a review of existing testing algorithms should be performed by diagnostic laboratories in order to avoid errors and that confirmatory testing should be considered (Australian Government Department of Health; McAuliffe et al., 2018). However, international guidelines differ in their recommendations around the need for this (Payne and Parashar). Further studies, particu- larly evaluating which patient- and season-specific factors influence the reliability of results, may help inform testing protocols for clinicians and laboratories at both a local and national level (Lopez-Lacort et al., 2016). In New Zealand, laboratories employ a range of tests, such as enzyme immunoassays (EIAs), immunochromatography (ICT), and polymerase chain reaction (PCR) to detect rotavirus antigen or nucleic acid. Prior to vaccine introduction, confirmatory testing was not performed prior to reporting of results, and rotavirus infections are not notifiable to pub- lic health authorities. Labtests (LTA) and Northland Pathology laboratories (NPL) serve as the sole community laboratories for an urban and rural population of 1.6 million (60% NZ European, 23% Asian, 15% Pacific, and 11% Maori) Diagnostic Microbiology and Infectious Disease xxx (xxxx) xxx–xxx ⁎ Corresponding author. Tel.: +6495747200. E-mail address: Gary.McAuliffe@labtests.co.nz (G.N. McAuliffe). DMB-14689; No of Pages 5 https://doi.org/10.1016/j.diagmicrobio.2018.10.003 0732-8893/© 2018 Elsevier Inc. All rights reserved. Contents lists available at ScienceDirect Diagnostic Microbiology and Infectious Disease journal homepage: www.elsevier.com/locate/diagmicrobio Please cite this article as: McAuliffe GN, et al, Suboptimal performance of rotavirus testing in a vaccinated community population should prompt laboratories to review their rotavi..., Diagn Microbiol Infect Dis (2018), https://doi.org/10.1016/j.diagmicrobio.2018.10.003