Validation of nested PCR-SSP typing. Recipient 01 (HLA-A type: A*0301, A*2402) was transfused with fresh blood from 2 healthy, HLA-typed volunteers (donor 1: A*0101; donor 2: A*2501, A*3002). The bar graph shows the results of amplification of 2 pretransfusion samples and a posttransfusion sample by primer mixes for the HLA-A locus alleles used in this system (Table 2). Analysis of both pretransfusion samples produces a similar pattern of amplification in which primer mixes 4 and 5 represent recipient alleles (A*0301) and (A*2402), respectively (dark shading). Potentially informative primer mixes are those that do not give rise to nonspecific products in any of the multiple tests of the pretransfusion samples. In this case, these are primer mixes 151, 9, 12, 152, 18, 184, and 154. Analysis of the posttransfusion samples shows additional amplification with primer mix 151, which is indicative of donor 1 (A*0101), and primer mixes 9 and 18, which are both indicative of donor 2 (A*2501 and A*3002, respectively; hatched shading). Primer mixes 3, 6, 10, 13, 14, 15, 153, 23, and 24 detect nonspecific products in the pretransfusion samples and thus are noninformative in the analysis of posttransfusion samples (light shading). The requirement for testing pretransfusion samples is demonstrated by the results with primer mix 10, which potentially should amplify A*2501 from donor 2. In this example, although there was amplification in the posttransfusion sample, the reaction has to be excluded from the analysis because of the nonspecific amplification present in the pretransfusion samples.