Methodologic Variations in the Determinations of Anti-Platelet Factor 4-Heparin Antibodies in Patients Suspected of Having Heparin Induced Thrombocytopenia: Diagnostic and Prognostic Implications.

It is now widely believed that the pathogenesis of heparin induced thrombocytopenia (HIT) is largely mediated via the generation of anti-platelet factor 4-heparin antibodies (APFA) in patients who are treated with heparin and related drugs. In addition antibodies to other heparin binding proteins such as neutrophil activating peptide-2 and interleukins also contribute to this syndrome and which are not detectable by the methods based on heparin platelet factor 4 capture probes. Currently, several immunologic methods mostly utilizing sandwich ELISA techniques to measure APFA in sera of patients with HIT syndrome are available. These methods utilize different capture probes including heparin platelet factor 4 complex (ASSERACHROM® HPIA; Diagnostica Stago, France), platelet factor 4 complexed to polyvinyl sulfonate (PF4 Enhanced; GTI, USA) and heparin protamine suflate along with platelet lysates as a source of PF4 (ZYMUTEST HIA IgGAM; Hyphen Biomedical, France). These different capture probes for the APFA have different affinity for the APFAs. Moreover, these capture probes can also bind to certain other heparin binding proteins. To compare these three methods, samples (n = 100) were selected from banked sera that had been referred to Loyola University Medical Center for the quantitation of HIT antibodies and ¹⁴C Serotonin Release Assay (SRA). All of these specimens were initially positive in the GTI PF4 Enhanced ELISA for the presence of anti-heparin platelet factor 4 antibodies with a broad range of antibody titers absorbance range (0.4 – 2.5). For the head to head comparison all of these samples were assayed using the three different methods. In the GTI method the reassayed samples showed 93 out of 100 positive (93%), in the HPIA test 79 out of 100 (79%) and in the ZYMUTEST 56 out of 100 (56%) were positive. Interestingly, the correlation coefficients also showed marked variation (GTI vs ZYMUTEST r² = 0.38; GTI vs Stago r² = 0.49; ZYMUTEST vs Stago r² = 0.67). The prevalence of positive samples was not consistent in the three tests used. This data clearly shows that each of the different ELISA methods exhibit different performance characteristics in binding to not only the APFA but other proteins, which may contribute to the higher false positive prevalence. Moreover, the positive/negative cutoff limits in these assays are arbitrarily set without due consideration of the antibody titer, which can also account for the observed differences. More interestingly, all of the samples positive in the SRA (n = 14) were positive in each of the three methods used with the exception of two samples, which were positive in the GTI, ASSERACHROM® HPIA and negative in the ZYMUTEST. These results clearly indicate the differences in the diagnostic efficacy of various commercially available tests and warrant clinical field trials to validate their reliability in the monitoring of APFA. Furthermore, the diagnostic reliability of these tests can be further improved by immunoglobulin subtyping and utilizing other capture probes with platelet factor 4 and related proteins, which complex with heparins.