A novel diagnostic algorithm for heparin-induced thrombocytopenia.

Abstract

While diagnostic algorithm using PF4-heparin enzyme-linked immunosorbent assay (ELISA) optical density (OD), and heparin neutralization assay (HNA), or 4T score have been proposed to replace serotonin-release assay (SRA) for heparin-induced thrombocytopenia (HIT), their performance against SRA is unclear. In this study, we proposed and validated the performance of a new algorithm combining PF4-heparin ELISA optical density (OD), HNA and 4T score against SRA for HIT diagnosis. Heparin neutralization assays were performed on specimens submitted for HIT testing with positive PF4-heparin ELISA from December 2015 to September 2017, which were separated into a "training" and a "validation" data set. 4T scores were calculated for ELISA positive cases. A total of 123 consecutive unique patient samples had positive PF4-heparin ELISA with also HNA data, SRA data, and 4T scores available. Compared to SRA, a "laboratory criteria" (ELISA OD≥1.4 and HNA≥70%) had a sensitivity of 88% (14/16) and specificity of 91% (42/46), and a "combined criteria" (4T score=8, or 4T score=6 or 7 and ELISA OD≥1.0, or 4T score=4 or 5 and ELISA OD≥2.0) had a sensitivity of 75% (12/16) and specificity of 98% (45/46) in the training data set. Laboratory and combined criteria had 90% (56/62) concordance rate. Importantly, for these concordant cases, the diagnostic specificity is 100% (46/46). Based on the data, a novel diagnostic algorithm combining these 2 criteria was proposed and validated prospectively. A novel algorithm has high diagnostic accuracy and potentially could eliminate the need for SRA testing in at least 90% patients with suspected HIT.