Backgrounds:

Febrile seizures are the most common cause of seizure in childhood.They occur in the presence of a fever (>38 degrees Celsius) in children aged between 6 months and 5 years. Viruses are well-described as the predominant causative agents and specific viruses have previously been implicated, including Human Herpesvirus-6 (HHV-6),Influenza A and B, Respiratory syncytial virus (RSV), Parainfluenza, Adenovirus, Rhinovirus and Enterovirus.

Recent innovations in viral respiratory molecular diagnostics allow multiple viruses to be tested simultaneously using multiplex polymerase chain reaction (PCR). SNOTWATCH is a de-identified ecologic analysis platform capturing population level molecular diagnostic results together with hospital and primary care encounters. This project aims to utilise PCR data to understand the relationship between respiratory virus circulation and febrile seizures at a population level. We have created a novel statistical model for assessing these relationships in both time and space.

Methods

Our ecological study assessed relationships between presentations of febrile seizures and nine respiratory viruses detected at the largest hospital network in Melbourne, Australia, from 2010-2019. Associations were studied temporally and spatiotemporally through mixed-effects Poisson regression analysis, using monthly incidence of febrile seizures and positive PCR tests.

Results:

Febrile seizure incidence peaked in June-September each year. Temporal analysis showed febrile seizures were significantly associated with Human metapneumovirus (1.19 RR), Influenza A (1.49 RR), Influenza B (1.33 RR) and RSV (1.52 RR) (Figure 1). Spatiotemporal analysis supported the association between febrile seizures and Influenza A, Influenza B and RSV (1.25, 1.12 and 1.20 RR respectively, p<0.0

Conclusions/Learning Points:

With over 90,000 PCR results and almost 5,000 febrile seizure presentations, our findings confirm the importance of understanding viral circulation patterns and their implications for paediatric health outcomes. Our statistical method may be used in predictive modelling to inform public health policy.