Human T-cell lymphotropic virus: A simulation model to estimate residual risk with universal leucoreduction and testing strategies in Canada.

Abstract

In Canada, transfusion transmission risk of Human T-cell lymphotropic virus -I/II (HTLV) is addressed by universal leucoreduction and universal antibody testing. We aimed to estimate the risk with the current policy, if testing only first-time donors and if testing were stopped. Monte Carlo simulation was employed to estimate the proportion of red cell concentrate, random donor platelet and apheresis platelet units that would be released into inventory in each scenario (10 billion donors each). The model estimated the number of HTLV-positive donations not intercepted by testing, randomly assigned the number of HTLV particles/100 leucocytes using proportions from published data and randomly selected a postleucoreduction leucocyte count from quality control data. Units were considered infectious if ≥9×104 copies of HTLV provirus. With universal leucoreduction in place, the residual risk of releasing an HTLV potentially infectious unit with universal testing was 1 in 1·2 billion units (0, 1 in 55·9 million), with testing only first-time donors 1 in 7·1 million (0, 1 in 1·05 million) and with no testing 1 in 1·0 million (0, 1 in 178600). The efficacy of leucoreduction was >99·5% (lower bound 95·7%) for all scenarios. With universal leucoreduction in place, switching from universal testing to testing first-time donors would incur very low risk.