The Residual Risk of Transfusion-Transmitted HIV Infection in a Public Brazilian Blood Center.

Estimating risks of transfusion-transmitted infections is essential for monitoring blood safety and for helping physicians and patients decide about the risks/benefits of an allogeneic transfusion. In 2000, the residual risk of HIV in our institution was estimated in 1/325,000 units transfused and the prevalence of HIV among all blood donors was 0.06%. The goal of this study was to calculate the current residual risk of transfusion-transmitted HIV infection in the Regional Blood Center of Ribeirão Preto, University of São Paulo, Brazil. We evaluated 292,587 repeated donations of a total of 403,568 donations between january/2002 and december/2005. During this period, we identified 43 seroconversions among the repeated donors. The serological tests performed were 2 different enzyme immunoassays (EIA): HIV 1/2 and a conjugated p24 antigen/antibody as screening tests and Western Blot as a confirmatory test for all positive samples. In order to calculate the residual risk, we used the incidence rate/window period (IR/WP) model, developed in parallel by investigators from the National Heart, Lung and Blood Institute-sponsored Retrovirus Epidemiology Donor Study (REDS), the CDC and the American Red Cross. The IR takes into account the number of incident cases (numerator), the total number of repeated donations and the mean interdonation interval length (multiplication of the last two entities provides the number of person-years at risk or denominator). The mean interdonation interval length can be substituted for the mean seroconversion interval length for all incident cases if the first one can not be accurately measured, so we used this alternative. Once the IR is known, we calculated the residual risk dividing the IR by 365.25 days and then multiplying by the WP of the test used (p24=15 days). Our results are: IR= 43 cases/292,587 repeated donations × 3.2 years = 4.59×10⁻⁵. The current residual risk is (IR/365.25 days) × 15 days= 1.5/million units transfused or 1/667,000 units. The prevalence of HIV among all blood donors in 2005 was 0.035%. Although we know that the correlation between prevalence and residual risk is not straightforward, they have a certain degree of correlation and they both have declined since 2000. During these last 5 years, we have made some modifications in our donor selection towards a more rigorous donor screening and since january/2002 we have introduced the p24 antigen/antibody EIA as part of the screening serological tests. We concluded that the more stringent donor risk factor screening and the introduction of the p24 antigen test have substantially reduced the residual risk of transfusion-transmitted HIV infection in our institution.