Abstract

Introduction: If HIT is considered, it should be confirmed or ruled out immediately to guarantee optimal (and low cost) anticoagulation or thrombosis prophylaxis for the patient. When diagnosis is not made, it may cause fatal thromboembolic disease in the patient due to enhanced platelet activation. The pathomechanism is well established: Antibody (ab) binding to heparin/PF4-complex (antigen) forms an immunocomplex. The complex binds to the Fc-Receptor of the platelet membrane resulting in platelet activation and further thrombin generation. Persistent exposure to heparinleads to continuous complex formation. To increase sensitivity and specificity, combination of an antigen test with a functional test is recommended: the antigen-test (EIA) detects ab binding (IgG, IgA und IgM) to heparin/PF4-complex or polyvinylsulfat-complex. Functional tests (HIPA-Test, Serotonin-release test or platelet aggregation studies) detect HIT- ab against different antigens - however, only IgG class. Therefore, the positive predictive value of a functional test is higher than for an antigen test to diagnose HIT. Ab of the IgA and IgM class have no platelet activating capacity and therefore, they are thought to be of minor clinical relevance.

Aim of the study: With HIT-IgG-EIA tests approaching the market, the question is, which EIA-test should be performed and will the result change our diagnosis made before by the conventional EIA together with the functional assay and complete clinical data.

Material and Methods: We compared the results of 28 out of 85 samples send over a one month period to our lab for HIT-testing. For this study positive or borderline positive samples were selected and only 5 of the negative’s. All samples we run the same day with the conventional EIA (GTI) and later with the ID-PaGIA gel-particle-card (IgG/IgM/IgA) (DIAMED Comp., Suisse) and two different EIA (only IgG) by GTI Diagnostics (Waukesha, WI) and Haemochrom Diagnostica (Essen, Germany).

Results: Out of the 28 samples, initially 5 were tested negative, 21 were tested positive (OD 0.5– 3.1, mean OD 1.41; positive if OD >0.4) and 2 being at the cut off. All 5 samples were tested negative in all assays, the 2 borderline samples were tested negative in the other assays. Of the 21 positive. samples 5 were negative by the ID-PaGIA gel-particle-card. Of the 21 positive. samples 13 were positive in the GTI-EIA (IgG only) and 7 in the Haemochrom EIA (IgG only) and 4 out of those had positive functional studies (platelet aggregation assay) - confirming the presence of IgG ab. However, the ID-PaGIA gel-particle-card missed 3 positive samples, when compared to the IgG-only EIAs.

Conclusion: As expected there were fewer positive samples in the EIA detecting only IgG class ab, however the comparability of those two EIAs was pure, surprisingly. Further investigation will be done to confirm this observation. Sensitivity and specificity of the ID-PaGIA gel-particle-card is lower then of the EIA. In our small series we could not find an OD-cut-off-point for positive EIA-results (IgG/M/A) in comparison with a positive or negative result for the IgG-EIA. In the future we might add an IgG-only EIA to the established GTI-EIA (IgG/M/A) to our work up and use the clinical scoring system (4T’s) vigorously before the diagnosis HIT will be made, to avoid overdiagnosing patients having positive results in the GTI-EIA (IgG/M/A) due to other underlying clinical conditions.

Author notes

Disclosure: No relevant conflicts of interest to declare.