Data of the German Haemovigilance System were collected from 1997 to 2007 and assessed on the basis of pre-defined safety standards. Suspected cases of serious adverse reactions following transfusions reported to the Paul-Ehrlich-Institut were evaluated on the basis of national criteria, and the definitions of International Society of Blood Transfusion (ISBT) in compliance with defined causality criteria. The suspected cases were rated as confirmed and unconfirmed transfusion reactions. Assessment of causality took into consideration the clinical course of the adverse reaction and, if necessary, information about donation and manufacturing. Of the 5128 suspected serious adverse reactions, 1603 could be confirmed. Referring to the absolute figures, acute transfusion reactions (e.g. allergic reactions, hypotension and dyspnoea) were recorded most frequently, followed by transfusion-related acute lung injury (TRALI), haemolytic reactions, transfusion-related bacterial infections and virus infections. The majority of the 52 transfusion-related fatalities (14 each) were due to TRALI and acute transfusion reactions (mostly severe allergic reactions). Referred to the blood products administered, immune TRALI cases and TRALI-related fatal courses were most frequently reported after administration of fresh frozen plasma (FFP) (15/10(6) and 3.5/10(6) units, respectively), transfusion-related bacterial infections after administration of platelet concentrates (7/10(6) units), acute haemolytic transfusion reactions after administration of red blood cell concentrates (2.3/10(6)units) and acute transfusion reactions after administration of red blood cell or platelet concentrates (7.8/10(6) and 13/10(6) units, respectively). Despite the high safety standard required for blood products in Germany, there is still room for reducing the frequency of isolated cases of transfusion reactions by targeted action.

Pigeons were trained in a conditional discrimination paradigm to differentiate successively presented visual arrays according to the relative number of their elements, Transfer tests with novel stimuli demonstrated that they discriminated the categories of "many" (6 or 7) from "few" (lor 2) items. In further tests, other new stimuli were introduced that consisted not only of these training numerosities, but also ofthe intervening ones (3, 4, and 5). Variationsin the birds' discrimination performance corresponded to the order of stimuli on a numerosity dimension. This serial ordering was maintained when other factors such as brightness, size, shape, area, and contour of the elements were systematically controlled across tests. Smaller numerosities were somewhat better discriminated than those at the higher end of this test range.

The comparison of the pre- and the post-implementation period demonstrated a significantly reduced risk of TRALI events comparing 2006/2007 with 2010 (P-value: <0·01). Furthermore, no case of TRALI-induced fatality occurred after the implementation of risk-minimization measures.

Methods: In order to evaluate the benefit of risk minimisation measures, reporting rates of transfusion-transmitted bacterial infections (TTBI) were calculated on the basis of annual reports and distributed blood components. Following the implementation of risk minimisation measures in 2003 and 2008, a comparison of pre- and post-implementation periods was performed. Results: During a period of 14 years, 90 cases of TTBI were confirmed, 34 were caused by red blood cell (RBC) concentrates, 5 by fresh frozen plasma, and 51 by platelet concentrates (PCs). The overall reporting frequency was 1 TTBI in 1.91 million RBC units; 1 TTBI in 0.094 million PC units, and 1 TTBI-associated fatality in 0.57 million PC units. From 2001–2004 the reporting rate was 13.7 per million PC units; 2005–2008, after the implementation of pre-donation sampling; it was 10.8 per million PC units (p > 0.5). After limitation of the shelf life (2008), the reporting rate decreased to 4.49 per million PC units (p = 0.12), and one case of related fatality was reported. Agents with low pathogenicity were reported in 14 of 41 immunosuppressed patients (34%) but only in 1 of 13 patients with non-haematological/oncological diseases. Conclusion: TTBI and associated fatalities could be gradually reduced by the risk minimisation measures, but further strategies such as implementation of sensitive screening tests or pathogen-reducing approaches should be discussed.

For three IVIGs, the drug-related TEE-reporting rates remained stable from 2006 to 2011 (0-0·83 cases per 1000 kg IVIG distributed). In contrast, the TEE rate of one IVIG increased significantly from 0·33 cases in 2006 to nearly nine cases in 2010 (P < 0·001). The NAPTT testing of IG products with a low TEE rate revealed a NAPTT time >200 s and a NAPTT ratio >0·8, whereas TEE-associated batches of IG products with an increased TEE rate had a NAPTT ratio <0·8. After modifications of manufacturing processes, a normalization of NAPTT results and a decrease in TEE rates could be demonstrated.