Heparin-induced thrombocytopenia (HIT), a prothrombotic drug reaction caused by platelet-activating anti-platelet factor 4 (PF4)/heparin antibodies, is commonly diagnosed with the platelet serotonin-release assay (SRA) or enzyme immunoassay (EIA). Researchers from McMaster University recently reported in the American Journal of Hematology that some patients may have SRA-negative HIT and require testing with a modified assay.
“Although few diagnostic tests can be expected to have 100% sensitivity, our group has traditionally regarded the SRA as having high sensitivity for HIT,” the authors, led by Theodore Warkentin, MD, Professor in Pathology and Molecular Medicine at McMaster University, wrote. “Recently, however, other investigators have proposed that SRA-negative HIT might be relatively common.”
This observation prompted Dr. Warkentin and his colleagues to evaluate the frequency of SRA-negative HIT in a cohort of 430 patients who were previously tested for HIT. These patients were part of a prospective study that evaluated the “4Ts” pretest scoring system and had residual serum and plasma samples available.
The researchers performed detailed clinical summaries of patients who were found to have HIT, including documentation of all heparin exposures, serial platelet counts, thrombotic events, and any other relevant events to explain thrombocytopenia.
The researchers identified a total of 35 patients who met criteria for SRA-positive HIT – all of whom had a clinical and laboratory picture compatible with HIT and also a positive EIA (≥1.00 optical density units). Most of these patients (n=20; 57% of the SRA-positive cohort) had objectively documented thrombosis (3 arterial events and 17 venous events) that occurred in association with the episode of HIT. The median platelet count nadir in this group was 44×109/L. Three (9%) of the SRA-positive patients died within 30 days of follow-up.
Another 27 patients were found to be EIA-positive/SRA-negative, defined as <20% serotonin-release and a positive EIA; these patients were evaluated further to see whether some cases were consistent with a strict diagnosis of SRA-negative HIT.
The degree of thrombocytopenia in this group was similar to that observed in SRA-positive patients, with a median platelet count nadir of 49×109/L. However, there were distinct differences in the frequency of thrombosis, which was higher in the SRA-positive group (20/35 [57%] vs. 4/27 [15%]; p=0.0013), while the 30-day mortality was higher in the SRA-negative patients compared with the SRA-positive patients (8/27 [30%] vs. 3/35 [9%]; p=0.045).
“We suggest that the greater mortality in SRA-negative/EIA-positive patients reflects the life-threatening nature of several of the underlying non-HIT explanations for thrombocytopenia,” including cancer, sepsis, and pneumonia, the researchers noted.
Of the SRA-negative group, only 1 patient met stringent criteria for SRA-negative HIT, including a clinical picture compatible with HIT (4Ts ≥4 points) and positive results on the “PF4-SRA,” a modified SRA using high concentrations of added PF4 rather than heparin. The patient was an 85-year-old woman who developed an otherwise-unexplained thrombocytopenia 7 days after starting heparin.
Only 2 of the remaining 26 EIA-positive/SRA-negative patients tested positive in the PF4-SRA for sub-threshold levels of platelet-activating antibodies – both of whom were considered to have a low probability of HIT (4Ts score ≤3).
For the patients in whom the PF4-SRA was negative, alternative explanations for thrombocytopenia were found. “Moreover, most of these patients were scored as low probability for HIT,” the investigators reported. “After the SRA test result returned negative, 10 of the 27 SRA-negative/EIA-positive patients received subsequent heparin re-exposure, with none of these 10 patients developing recurrent thrombocytopenia attributable to HIT.”
Overall, the conventional SRA was shown to have a sensitivity of 97%, and although the study supports the concept of SRA-negative HIT, “the frequency of this phenomenon appears to be low,” the authors concluded. “[These results] also provide a rational approach to investigate patients who could have SRA-negative HIT, namely by performing the PF4-SRA in an appropriate patient who has a clinical picture suggestive of HIT and in whom an EIA yields a result of ≥1.0 units.”
The study is limited by its reliance on data from a single medical community (McMaster University in Hamilton, Ontario), and the authors acknowledged that they tested stored serum samples that were up to 10 years old, which may have led to sample degradation and the underestimation of SRA-negative HIT cases.
The authors report no relevant conflicts of interest.
Warkentin TE, Nazy I, Sheppard JI, et al. Serotonin-release assay-negative heparin-induced thrombocytopenia. Am J Hematol. 2020;95:38-47.