The objectives of this study were to demonstrate that the risk of PRRSV-contaminated aerosols entering a facility via retrograde air is a true risk though unfiltered points (i.e. idle fans); to titrate the minimum air speed necessary to introduce PRRSV-contaminated aerosols via retrograde air; and to validate commercially available interventions that have been designed to prevent this risk.

The study was conducted at the UMN SDEC production region model using an empty facility negative ventilated. One of the 2 fans was intentionally stopped while the other continued to operate. In order to measure the air speed of the retrograde air through the idle fan needed to transfer PRRSV (retrograde air titration), a common plastic shutter was challenged at various fan stages using 10 replicates of different PRRSV concentrations each (1 to 7 logs of the virus) in a liter which were generated using a cold-fog mister located on the exterior of the facility. To titrate the air speed needed to transfer PRRSV, a cyclonic collector was placed inside the facility. The measurements of retrograde air speeds and static pressures were collected for each fan stage. Treatments evaluated included the standard plastic shutter, a plastic shutter plus a canvas cover, a nylon windsock, an aluminum shutter plus a windsock and, a double shutter system (aluminum and plastic shutters). All 5 treatments were challenged as described in order to determine whether aerosolized PRRSV could penetrate the different treatments.

 
The results of this study suggest that a real risk of PRRSV entry may exists when there is a minimum retrograde air speed of 0.76 m/s. As well this study suggests that the plastic shutter and canvas cover do not offer complete protection against retrograde air movement and the risk of aerosolized PRRSV entry.
Results from this study indicate that retrograde air movement is a risk for PRRSV introduction in filtered farms, that it requires a minimum velocity of air flow and that not all interventions designed to reduce this risk are effective.