A Novel Test for the Rapid Rule Out of Heparin-Induced Thrombocytopenia Diagnosis in Intensive Care unit patients

Diagnosis of heparin-induced thrombocytopenia (HIT) in intensive care unit (ICU) patients represents a major challenge mainly because both the use of unfractionated heparin and the presence of thrombocytopenia are quite common. Despite the existence of several laboratory tests, accurate and prompt HIT diagnostics remains difficult. The ideal combination of an immunological and a functional test is restrictedto specialized laboratories, due to the complexity of the latter. We are in need of an easy-to-perform, widely accessible, rapid and reliably assay.

To prospectively evaluate the performance of the latelar-flow immunoassay STic HIT Expert^® (Diagnostica Stago, France) for the detection in ICU patients suspected for HIT.

Seventy two patients (40 males/32 females) hospitalized in ICU from January to June 2012 were included. Thirty one patients presented with sepsis, 27 underwent extracorporeal circulation (ECC), 21 were hemodialysed and 3 patients were receiving chemotherapy. Sixty one patients were treated with unfractionated heparin and 11 patients received low molecular weight heparin (LMWH). A 4T's score was performed for all patients. All samples were tested in polyspecific ELISA (Zymutest Hyphen Biomed, Neuville-Sur-Oise, France), STic HIT Expert^® (Diagnostica Stago, France) and serotonin release assay. In case of a positive polyspecific ELISA, IgG, IgM et IgA isotypes were also performed. Sensitivity, specificity, positive and negative predictive values (PPV an NPV) of STic HIT Expert^® were determined against SRA.

All three tests (polyspecific ELISA, STic HIT Expert^®, SRA) were negative in forty patients and had a low HIT suspicion (4T's score: 0–4). In 10 out of 72 patients polyspecific all immunological tests and SRA were positive and HIT suspicion was intermediate or high (4T's score: 4–7). In 9 patients, ELISA tests and STic HIT Expert^® were positive but SRA was negative. These patients had a low HIT suspicion (4T's score: 1–4) and underwent ECC (6 out of 9), were hemodialyzed (3 out of 9) or complicated by sepsis (2 out of 9). On the other hand 13 out of 72 patients had ELISA tests positive but STic HIT Expert^® and SRA negative. The prevalence of sepsis was high in these patients (8 out of 13), 3 patients underwent ECC and one patient was hemodialysed. STic HIT Expert^®, polyspecific and IgG specific ELISA had an excellent sensitivity and negative predictive value at 100%. Moreover STic HIT Expert^® was associated with a smaller number of false positive results than the ELISAs. (Table)

STic HIT Expert^® has an excellent performance with a high negative predictive value (100%) and a satisfactory specificity (85%). Less false positive results are detected with STic HIT Expert^® than with polyspecific and IgG specific ELISAs. Moreover the test offers a shorter turnaround than ELISA tests and is an easy-to-use single sample test. These characteristics could help avoid HIT over diagnosis in ICU patients.

No relevant conflicts of interest to declare.