Influence of different batch-rearing systems on swine influenza virus (SIV) spread and persistence in farrow-to- finish farms using a stochastic modeling approach

International audienceSwine Influenza, formerly described as an episodic disease, has been shown to persist in an enzootic form in farrow-to-finish farms. Previous studies on commercial farms showed that Swine Influenza A Virus (SIV) infection generally affects weaned piglets in nursery at a particular age and occurs on each successive batch. However, the transmission process between batches is still not well understood. The aim of this study was to analyze the conditions for persistence and recurrence of SIV in farrow-to-finish pig herds reared with different batch-rearing systems (4-, 5-, 7-, 10- and 20-batches).The within-farm dynamics of SIV was modeled using a stochastic metapopulation compartmental model. A population dynamics model, accounting for different batch-rearing systems, was coupled with a SIV epidemiological model. Two subpopulations were considered, corresponding to breeding sows and growing pigs respectively, and interacting during lactating stage. The infectious process was represented by batch-specific SIRS models, to which were added different compartments accounting for the impact of MDAs on piglets infection (partial protection and impairment of the post-infectious immune response). Moreover, an indirect between-batch transmission rate was considered. For each batch system, an infected gilt was introduced in batch 1 in insemination room. Gestating sows were housed in a large dynamic group. The model has been used to evaluate the minimal conditions for virus transmission between batches and persistence at the population level. The impact of batch-rearing systems was also evaluated.Whatever the batch system of the farms, the introduction of one infected gilt in insemination room resulted in SIV persistence for several months in pig population. Relatively low transmission rate between animals from different batches was found sufficient to produce recurrent infections at fixed age in growing pigs population, a close to field observations. The batch system was shown as a pivotal factor favoring the spread and persistence when short time-intervals were implemented between successive batches.The introduction of an infected gilt in the breeding sows population reproduced the same dynamic in this modeling approach as observed in commercial farms (i.e. infection of consecutive batches of piglets in nursery). The results of the model showed that a lack of internal biosecurity between batches in the nursery part could partially explain the propagation and maintenance of the virus from batch to batch. Moreover, several batch-rearing systems were found more at risk for SIV persistence at the metapopulation level