Monitoring heparin therapy: relationships between the activated partial thromboplastin time and heparin assays based on ex-vivo heparin samples

Five different APTT-reagents, two amidolytic anti-IIa assays, one amidolytic anti-Xa assay, and one coagulometric anti-Xa/anti-IIa assay were used to assess the effect of heparin in patients treated for venous thromboembolic disease. Good correlations were observed between log-transformed APTTs determined with the various reagents (correlation coefficients: 0.92-0.96). Nevertheless there were important differences in the slopes of the lines of relationship between the APTT reagents. Good correlations were observed between the anti-Xa and anti-IIa assay results (correlation coefficients: 0.92-0.97). However, the amidolytic anti-Xa activity was significantly higher (p less than 0.001) than the two amidolytic anti-IIa activities. Less good correlations were observed between the log-transformed APTTs and the anti-Xa or anti-IIa activities (correlation coefficients: 0.64-0.78). The correlations were improved by transforming the APTT into APTT-ratio, i.e. the ratio of the patient's APTT to the same patient's APTT after removal of heparin from the plasma sample by means of ECTEOLA-cellulose treatment. The correlation coefficients of log (APTT-ratio) with anti-Xa or anti-IIa ranged from 0.76 to 0.87. For both APTT and amidolytic heparin assay, the response to in vitro heparin was different from the response to ex vivo heparin. Therefore, equivalent therapeutic ranges should be assessed by using ex vivo samples rather than in vitro heparin. Because of the response differences between the APTT reagents, it is not adequate to define a therapeutic range for heparin therapy without specification of the reagent.