The GTI QuikScreen test is an enzyme-linked immunosorbent assay (ELISA) that uses soluble HLA class I antigens as targets. In tests of 5,893 human serum specimens, we evaluated the reliability, sensitivity, and utility of the GTI QuikScreen test for detecting HLA class I-specific antibody. We found that the test could reliably detect HLA-specific antibodies of the immunoglobulin G (IgG) but not the IgM class. The degree of correlation with lymphocytotoxicity testing varied among the different serum sources, with the best correlation achieved with sera from renal transplant candidates (r > 0.7) and the poorest with sera from patients with end-stage liver disease (r ‫؍‬ 0.26), possibly because of elevated alkaline phosphatase levels in the liver patients. Test reproducibility was high (96%), and test failure rate was low (1.7%). The test sensitivity is comparable to that of the antiglobulin cytotoxicity and, possibly, even flow cytometric tests. There was a highly significant (P < 0.001) correlation between the optical densities obtained in the ELISA and the percent panel reactive antibody determined by cytotoxicity testing. Therefore, although designed only to determine the presence or absence of HLA-specific antibody, GTI QuikScreen test results also provided an indication of the extent of sensitization. The test is one of the most effective and efficient ways to determine if antibodies producing a positive result in crossmatch tests are specific for HLA class I antigens. As an adjunct to serum screening by cytotoxicity testing, the GTI QuikScreen test can produce a substantial savings of time and effort that reduces the cost to the laboratory and to the patient.

The random-amplified polymorphic DNA (RAPD) assay was used to generate DNA fingerprints for 16 isolates of "Haemophilus somnus," and one isolate each of "Haemophilus agni," "Histophilus ovis," "ActinobaciUlus seminis," Pasteurella haemolytica, and Escherichia coli. The RAPD assay differentiated among "H. somnus" isolates, which shared similarity coefficients of 0.46 to 1.00 on the basis of pairwise comparisons of RAPD markers produced with nine random decamer primers. Three virulent encephalitic "H. somnus" isolates exhibited identical banding patterns, suggesting a common clonal ancestry. The RAPD assay clearly distinguished between the "H. somnus"-"H. agni"-"H. ovis" group and the other bacterial species tested. The results of the present study suggest that DNA fingerprinting of "H. somnus" isolates by the RAPD assay could be valuable in revealing subspecific divisions within this largely unexplored species.