Background: The risk of infection with transfusion-transmitted viruses has been reduced remarkably. A zero-risk blood supply, however, remains a popular goal. A 3-year prospective donor study was conducted in the Epirus region of Greece to determine the prevalence of human immunodeficiency virus (HIV), human T-lymphotropic virus (HTLV), hepatitis B virus, and hepatitis C virus (HCV). Herein, we report the prevalence of HIV, HTLV, and HCV infection markers in this area. Methodology: Between January 1, 1995 and December 31, 1997, 6696 donors were investigated for the presence of anti-HIV, anti-HTLV, and anti-HCV antibodies using standard enzyme immunoassays (EIA). Every sample with anti-HCV reactivity by third-generation EIA was further investigated using a third-generation recombinant immunoblot assay (RIBA 3.0) and HCV-RNA by a combination of polymerase chain reaction (PCR) and DNA EIA. Results: None of the donors tested positive for anti-HIV or anti-HTLV antibodies. In contrast, anti-HCV was detected in 41 donors (0.61%). Using a RIBA 3.0 test, eight donors tested positive and eight had indeterminate results, while 25 tested negative. Seven of the eight donors with both EIA and RIBA 3.0 reactivity had increased levels of aminotransferases and detectable serum HCV-RNA. The remaining 34 donors had repeatedly normal aminotransferases and three times negative HCV-RNA. Liver biopsy was performed in anti-HCV/HCV-RNA-positive donors (7/41). The lesions were compatible with chronic hepatitis C in all of them. Conclusion: A zero prevalence of HIV and HTLV infection markers was found. Although the number of annual donations in this study was relatively low, the negative data for HIV and HTLV clearly indicate that rates of these infections are low in our region and that infected donors will be seen infrequently. HCV infection in blood donors remains very low in our region and is similar to the data reported in other industrialized countries. In fact, the prevalence of definite HCV infection seems to be very low (7/6696; 0.1%). However, a significant proportion of anti-HCV-reactive donors by third-generation EIA (33/41) had indeterminate or negative results by the RIBA 3.0. The latter donors were repeatedly negative for HCV-RNA. This finding may indicate that some donors tested false-positive for anti-HCV, although the possibility of true HCV infection contracted in the distant past cannot be excluded. In our opinion, close attention to mandatory principles of transfusion medicine, along with the screening of plasma donors using nucleic acid amplification technology, are the only methods that can further ensure the safety of our blood supply.