Background: Heparin-induced thrombocytopenia (HIT) is a potentially life-threatening adverse drug reaction with numerous diagnostic challenges. Diagnosis of HIT begins with 4T score clinical assessment, followed by laboratory testing for those not deemed low risk. Laboratory testing for HIT includes screening [enzyme-linked immunosorbent assay (ELISA)] and confirmatory [serotonin release assay (SRA)] assays, wherein SRA testing can be pursued following a positive ELISA result. These tests aid diagnosis of HIT, but also introduce interpretive challenges, additional costs, and delays in clinical intervention.
Methods: A retrospective review of 1011 HIT ELISA and 169 SRA tests performed over 5 years was conducted. ELISA percent inhibition and ELISA low-heparin optical density (OD) were evaluated for positive predictive value (PPV). Based on these findings, HIT ELISA reporting and management algorithm changes were implemented and metrics compared for 5 months pre- and post-intervention to assess intervention success.
Results: Equivocal and positive HIT ELISA interpretation showed poor PPV based on percent inhibition (0.20 and 0.32, respectively). However, rising low-heparin OD correlated with increasing PPV (PPV of 0.00 for OD values 0.40-1.00, 0.29 for values 1.00-1.99, and 0.91 for values >2.00). Data-driven intervention decreased ELISA positivity rates (13% to 5%), decreased rates of SRA confirmatory testing overall (13% to 9%), decreased SRA testing rates for patients with non-negative ELISAs (78% to 43%), and increased heparin resumption (20% to 57%). Hematology consults remained relatively stable (78% and 86%).
Conclusions: Low-heparin OD-based HIT ELISA interpretation yielded enhanced PPV compared with percent inhibition-based interpretation. Implementation of data-driven changes improved testing stewardship and clinical management for patients with non-negative ELISAs.
© Association for Diagnostics & Laboratory Medicine 2025.