Epidemiology of HIV, HBV and HCV and the Safety of Blood Products. Concerning Stark K, et al: Infections with HIV, HBV and HCV among blood donors in Germany 1998 and 1999. Infus Ther Transfus Med 2002;29:305–307 / Authors' Reply

In table 1 of their recent contribution Stark and co-workers list the number of donations from infected first-time and repeat donors reported to the Robert Koch-Institut, divided according to the different viruses (HIV, HBV, and HCV), to whole-blood and plasmapheresis donations and to the two time periods second half of 1998 and 1999. From these data, it should be possible to draw conclusions with respect to prevalence (first-time donors) and incidence (repeat donors) of the three infections among whole-blood and plasmapheresis donors. However, in the accompanying text these conclusions were either missing altogether or were incomplete. 1) Among first-time donors, each donation counted in the table corresponds to one donor. The rate of marker-positive donations may be interpreted as prevalence of infection in donor applicants/candidates for the respective donations (whole blood versus plasmapheresis) Among whole-blood donors this prevalence remains approximately constant for all three viruses from the second half of 1998 to the year 1999. Among plasmapheresis donors, however, there is a dramatic increase from 2/6,921 to 39/9,900 reported infections. Either 10 times as many infected persons chose to present for plasmapheresis donation in 1999 than in the second half of 1998 or not all infected first-time donors were reported to the Robert KochInstitut in 1998. In favor of the latter possibility is the fact that the Advisory Group ‘Blood’ (Arbeitskreis Blut), in its recommendation (votum 24), expressly requires reporting of confirmation of suspected infection after repeatedly reactive results in screening tests also for first-time donations. 2) According to table 1, in 1999 plasmapheresis donors have a significantly higher prevalence of HCV than whole-blood donors, but this is neither mentioned nor discussed in the text. 3) There is a contradiction between the table’s information on HBV prevalence in 1998 and 1999 in whole-blood donors and the comment in the text. According to table 1, there is a slight increase from 146.9/100,000 donations in the second half of 1998 to 150.1/100,000 donations in 1999. The text, however, reports a significant decrease between 1998 and 1999 with a p value of 0.04. 4) In addition, Stark et al. listed the rate of infectious donations of repeat whole-blood and plasmapheresis donors. Conclusions on the incidence of the infections in the respective donor pools can only be drawn if it is known from how many donors these donations are derived. A mathematical model exists which allows calculation of the residual risk from the incidence of the infection in the donor population and the interval between donations. Stark and co-workers did not do the calculations ‘... because the data on the time intervals between repeat donations were incomplete or inconsistent in a high proportion of the reporting services.’ Even if the poor quality of their data did not allow the calculation, Stark and co-workers could (and should perhaps) have mentioned that plasmapheresis donors generally donate much more frequently than (10 times as frequently?) whole-blood donors. The lower rate of HCV-infectious donations from repeat plasmapheresis donors consequently corresponds to a higher incidence of HCV infections in that donor population. 5) The last sentence states that ‘the data on infections among blood donors demonstrate the high safety of blood products in Germany and the benefit of careful donor selection.’ Unfortunately, the first point is not adequately discussed to clearly substantiate this indisputable observation. First-time donors who test positive do not pose a risk with respect to the infections for which they are tested. More problematic are the significant differences in prevalence between collectives if one assumes that an increased prevalence of known parenterally transmitted infections at some point in the future may be related to an increased prevalence of a still unknown parenterally transmitted infection. HIV was an example of this in the past. Prevalence of a virus infection in a donor group only allows to draw limited conclusions with respect to the much more important incidence in that donor group. Nevertheless, it is remarkable that not only the prevalence, but very likely also the incidence of HBV and HCV infections, is higher in plasmapheresis than in whole-blood donors. This may be irrelevant for the safety of plasma derivatives due to NAT testing and virus inactivation procedures. Quarantine should maintain the safety of noninactivated plasma for therapeutic purposes at least with respect to HCV. Problems could possibly arise if plasmapheresis donors are also requested to donate cellular blood products. It would be of utmost interest to analyze prevalence and incidence of the relevant virus infections in thrombapheresis donors separately as the characteristics of thrombapheresis donation Letters to the Editors · Briefe an die Herausgeber