SAN DIEGO – Both the 4Ts Score and the HIT Expert Probability (HEP) Score are useful in clinical practice for the diagnosis of heparin-induced thrombocytopenia, but the HEP score may have better operative characteristics in ICU patients, results from a “real world” analysis showed.
“The diagnosis of heparin-induced thrombocytopenia (HIT) is challenging,” Allyson M. Pishko, MD, one of the study authors, said at the biennial summit of the Thrombosis & Hemostasis Societies of North America. “The 4Ts Score is commonly used, but limitations include its low positive predictive value and significant interobserver variability.”
One external prospective study showed operating characteristics similar to those of 4Ts scores ().
The aim of the currentwas to validate the HEP Score in a “real world” setting and to compare the performance of the HEP Score versus the 4Ts Score. The researchers enrolled 292 adults with suspected acute HIT who were hospitalized at the University of Pennsylvania or affiliated community hospitals, and who had HIT laboratory testing ordered.
The HEP Score and the 4Ts Score were calculated by a member of the clinical team and were completed prior to return of the HIT lab test result. The majority of scorers (62%) were hematology fellows, followed by attendings (35%), and residents/students (3%). All patients underwent testing with an HIT ELISA and serotonin-release assay (SRA). Patients in whom the optical density of the ELISA was less than 0.4 units were classified as not having HIT. The researchers used the Wilcoxon rank-sum test to compare HEP and 4Ts Scores in patients with and without HIT.
Of the 292 patients, 209 were HIT negative and 83 had their data reviewed by an expert panel. Of these 83 patients, 40 were found to be HIT negative and 43 were HIT positive, and their mean ages were 65 years and 63 years, respectively. Among the cases found to be positive for HIT, 93% had HIT ELISA optical density of 1 or greater and 69.7% were SRA positive. The median HEP Score in patients with and without HIT was 8 versus 5 (P less than .0001).