Evaluation of a new automated panel of assays for the detection of anti-PF4/heparin antibodies in patients suspected of having heparin-induced thrombocytopenia.
Thromb Haemost. 2010 Jun 10;104(2)
Authors: Legnani C, Cini M, Pili C, Boggian O, Frascaro M, Palareti G
Heparin-induced thrombocytopenia (HIT) is a life-threatening complication of heparin treatment; the prognosis depends on early and accurate diagnosis, and prompt start of alternative anticoagulants. Because of high sensitivity, the commercially available immunologic assays are widely used, though not suited to be run on single samples and with a turnaround time of 2-3 hours. We evaluated two new, rapid, automated, semi-quantitative chemiluminescent immunoassays in HIT suspected patients: HemosIL(R) AcuStar HIT-IgG(PF4-H) (specific for IgG anti-PF4/heparin antibodies) and HemosIL(R) AcuStar HIT-Ab(PF4-H) (detecting IgG, IgM and IgA anti-PF4/heparin antibodies) (both from Instrumentation Laboratory). A total of 102 patients with suspected HIT were included; HIT was diagnosed in 17 (16.7%). No false negative cases were observed using either the HemosIL AcuStar HIT-IgG(PF4-H) or the HIT-Ab(PF4-H) assay (sensitivity and negative predictive values = 100%; negative likelihood ratios <0.01). The specificity was higher for the HemosIL AcuStar HIT-IgG(PF4-H) in comparison with that of the HemosIL AcuStar HIT-Ab(PF4-H) (96.5% vs. 81.2%). Higher values of the HemosIL AcuStar HIT-IgG(PF4-H) were associated with increased probability of HIT. Patients with confirmed HIT and thrombotic complications had significantly higher levels of HemosIL AcuStar HIT-IgG(PF4-H) than those without thrombotic complications. The HemosIL AcuStar HIT-IgG(PF4-H) and HIT-Ab(PF4-H) assays showed a very high sensitivity, and therefore they can reliably be used to rule out HIT in suspected patients. The diagnostic specificity was greatly increased by using the HemosIL AcuStar HIT-IgG(PF4-H). Both the assays are reproducible (CVs <6%), rapid (turnaround time 30 minutes), automated, and semi-quantitative, and they can be run for single sample testing.
PMID: 20539902 [PubMed - as supplied by publisher]