Modeling the parvovirus B19 blood safety risk in Australia

BACKGROUND Three probable cases of transfusion-transmitted (TT) parvovirus B19 (B19V) occurred in Australia between 2014 and 2017. This study aimed to determine the B19V DNA prevalence among blood donors, to model the risk to recipients of fresh components, and to assess risk management options. STUDY DESIGN AND METHODS Plasma samples from 4232 donors were tested for B19V DNA by polymerase chain reaction. Reactive samples were confirmed and viral load determined. A transmission-risk model was used to estimate recipient risk, and the risk from community exposure was estimated using seroprevalence data. RESULTS Two samples (0.0473%, 95% confidence interval [CI] 0.0130-0.172) confirmed positive for B19V DNA had a potentially infectious viral load of 105 IU/mL or higher. The estimated risk of a TT-B19V-associated significant complication was low overall at approximately 1 in 300,000 (95% CI, 1 in 82,000 to 1 in 1 million) fresh components transfused, with 3.1 (95% CI, 0.85-11.3) complications modeled per year. Among vulnerable recipient groups, the risk was higher than 1 in 15,000 patients, but the risk from community exposure far exceeded the transfusion risk for all patient and age groups. CONCLUSION In the context of the small contribution of transfusion to the burden of B19V disease, the significant costs that would be incurred by any strategy to reduce the risk, and given the significant uncertainties and likely overestimation of the risk, we conclude TT-B19V is a tolerable risk to blood safety, despite being high for some vulnerable recipient groups.