A prospective, blinded study of a PF4-dependent assay for HIT diagnosis.

Like Comment
Heparin-induced thrombocytopenia (HIT) is a life-threatening, pro-thrombotic, antibody-mediated disorder. To maximize the likelihood of recovery, early and accurate diagnosis is critical. Widely available HIT assays such as the Platelet Factor 4-Heparin ELISAs lack specificity, and the "gold standard" C14-labeled serotonin release assay (SRA) is of limited value for early patient management due to availability only through reference laboratories. Recent studies demonstrate that "pathogenic" HIT antibodies selectively activate PF4-treated platelets and that a technically simpler assay, the PF4-dependent P-selectin Expression Assay (PEA), may provide an option for rapid and conclusive results. Four hundred and nine consecutive adults suspected of HIT were classified as disease-positive, -negative or -indeterminate based upon predefined criteria that combined 4Ts scores and HIT ELISA results. Patients deemed "HIT-indeterminate" were considered disease-negative in the primary analysis and disease-positive in a sensitivity analysis. The ability of PEA and SRA to identify patients judged to have HIT was compared using receiver operating characteristic curve statistics. Using these predefined criteria, the diagnostic accuracy of PEA was high (Area under the curve [AUC] of 0.94; 0.87-1.0, 95% CI) and similar to that of SRA (0.91; 0.82-1.0, 95% CI). In sensitivity analysis, the AUCs of PEA and SRA were also similar at 0.88 (0.78-0.98, 95% CI) and 0.86 (0.77-0.96, 95% CI), respectively. The PEA, a technically simple non-radioactive assay that uses ~20-fold fewer platelets compared to the SRA had high accuracy for diagnosing HIT. Widespread use of the PEA may facilitate timely and more effective management of patients with suspected HIT.


Get PDF with LibKey


View the article @ Blood (sign-in may be required)

ClinOwl

The wider, wiser view for healthcare professionals. ClinOwl signposts the latest clinical content from over 100 leading medical journals.
2947 Contributions
1 Followers
0 Following

No comments yet.