Premium
Variability of anti‐PF4/heparin antibody results obtained by the rapid testing system ID‐H/PF4‐PaGIA
Author(s) -
SCHNEITER S.,
COLUCCI G.,
SULZER I.,
BARIZZI G.,
LÄMMLE B.,
ALBERIO L.
Publication year - 2009
Publication title -
journal of thrombosis and haemostasis
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.947
H-Index - 178
eISSN - 1538-7836
pISSN - 1538-7933
DOI - 10.1111/j.1538-7836.2009.03507.x
Subject(s) - heparin , heparin induced thrombocytopenia , platelet factor 4 , medicine , antibody , gastroenterology , immunology , chemistry
Summary. Background : Recent studies have shown that a low clinical pretest probability may be adequate for excluding heparin‐induced thrombocytopenia. However, for patients with intermediate or high pretest probability, laboratory testing is essential for confirming or refuting the diagnosis. Rapid assessment of anti‐PF4/heparin‐antibodies may assist clinical decision‐making. Objectives : To evaluate the performance of rapid ID‐H/PF4‐PaGIA. In particular, we verified reproducibility of results between plasma and serum specimens, between fresh and frozen samples, and between different ID‐H/PF4‐polymer lots (polystyrene beads coated with heparin/PF4‐complexes). Patients/Methods : The samples studied were 1376 plasma and 914 corresponding serum samples from patients investigated for suspected heparin‐induced thrombocytopenia between January 2000 and October 2008. Anti‐PF4/heparin‐antibodies were assessed by ID‐H/PF4‐PaGIA, commercially available ELISAs and heparin‐induced platelet aggregation test. Results : Among 914 paired plasma/serum samples we noted discordant results (negative vs. low‐titre positive) in nine instances (1%; 95%CI, 0.4–1.6%). Overall, agreement between titres assessed in plasma vs. serum was highly significant (Spearman correlation coefficient, 0.975; P < 0.0001). Forty‐seven samples tested both fresh and after freezing/thawing showed a good agreement, with one discordant positive/negative result (Spearman correlation coefficient, 0.970; P < 0.0001). Among 1376 plasma samples we noted a strikingly variable incidence of false negative results (none – 82%; 95%CI, 66–98%), depending on the employed ID‐H/PF4‐polymer lot. Faulty lots can be recognized by titrating commercial positive controls and stored samples of HIT‐patients. Conclusion : Laboratories performing the assay should implement stringent internal quality controls in order to recognize potentially faulty ID‐H/PF4‐polymer lots, thus avoiding false negative results.