Heparin-induced thrombocytopenia (HIT) diagnosis requires an integrated diagnostic approach combining clinical information with concomitant screening and confirmatory laboratory tests. First, we analyzed the test results and demographic data of patients with suspected HIT tested between 2015 and 2020. The observed parameters are comparable with descriptions in the foreign literature. Further, we have focused on the potential to improve our in-house confirmatory flow cytometric platelet-activating test, which tests native patient serum and 4 donor platelets in platelet-rich plasma (PRP). We tested different variables: number of platelet donors, preparation of donor platelets, and patient serum. The response of donor platelets to HIT antibodies in a patient serum was evaluated with an expanded panel of 10 donor platelets. We demonstrated that samples with a higher optical density value measured in the screening test (ELISA OD) or with a higher clinical probability of HIT (4T score) activated a higher number of donor platelets on average. Based on the results obtained and our experience with the functional test to date, we concluded that the use of 4 donor platelets is appropriate. The observed pre-analytical variables were also platelet washing with different buffers and heat inactivation of patient serum. The influence of these additional procedures on the test result was evaluated by testing patient samples in parallel according to the standard procedure with platelets obtained from the same donors, which has not been reported in the literature for this type of test. It seems that washed platelets vs. PRP and heat inactivation vs. native serum are not an improvement of the procedure. In our testing settings, heat inactivation could potentially resolve unclear cases as a parallel method to the standard protocol after adjusting the algorithm for result interpretation. We have designed a platelet activation index to help with result interpretation and assess platelet activation. Index has a very good negative predictive value at 1.1 or less and a good positive predictive value at 1.6 or more.
|