Monitoring unfractionated heparin therapy: Lack of standardization of anti‐Xa activity reagents

Introduction One of the main advantages of using anti‐Xa instead of activated partial thromboplastin time in monitoring of unfractionated heparin (UFH) therapy relies on its hypothesized standardization, with a unique therapeutic range defined to be 0.30 to 0.70 IU/mL. The aim of the present study was to compare the inter‐reagent agreement of anti‐Xa activity. Methods Citrate tubes were obtained from 104 inpatients on UFH. Plasma samples were stored frozen in aliquots at −70°C before being shipped to three accredited coagulation laboratories to be evaluated for anti‐Xa activity using their routine assay(s). Pooled normal plasmas spiked with dilutions of the 6th International Standard of UFH to achieve anti‐Xa activities up to 1.0 IU/mL were evaluated using the same techniques. Results In the plasmas from patients on UFH, the median anti‐Xa activity ranged from 0.37 IU/mL with one reagent to 0.57 IU/mL with another; results were in between (0.45 IU/mL) using two other reagents. Comparisons of results obtained using the different reagents demonstrated unacceptable bias up to 0.24 IU/mL between some reagents (41% difference). The concordance as whether anti‐Xa activities measured using different reagents were within or outside the therapeutic range was between 0.411 and 0.939 (kappa). Similar discrepancy was demonstrated for anti‐Xa activities when evaluating normal plasma spiked with the International Standard. A discrepancy of the same order of magnitude was demonstrated in the 2017 External Quality Assessment Program provided by the External Quality Control in Assays and Tests exercises. Conclusions The reported discrepancy between test results obtained using different anti‐Xa assays clearly suggests a lack of standardization of that assay with potentially significant impact on the patients’ anticoagulation.