Agreement between ACT and aPTT during extracorporeal membrane oxygenation shows intra- and inter-individual variation

We explored the relationship between activated clotting time (ACT) and activated partial thromboplastin time (aPTT) when used to monitor anticoagulation in patients undergoing extracorporeal membrane oxygenation (ECMO) support.

Data obtained in patients undergoing ECMO support between October 2012 and August 2013 in a single centre were reviewed. Clinical data were extracted from our Clinical Information System and ECMO database. ACT and aPTT values were paired when taken from the same patient, with the ACT preceding the aPTT and the heparin infusion rate was kept constant between samples. The aPTT and ACT were normalized by dividing by the mean of their respective reference ranges and are referred to as APR and N-ACT, respectively. Bivariate analysis and Bland-Altman plots were used to assess correlation and agreement. Mixed effects regression was used to model the effects of variables, including platelet count, creatinine and urea levels, plasma free haemoglobin, white cell count and ECMO flow rate on concordance between APR and N-ACT measurements.

The Pearson product-moment correlation coefficient in 15 patients was calculated as r=0.55. The Bland-Altman plot shows a mean difference between the APR and the N-ACT of -0.08. The 95% limits of agreement were -0.67 to 0.51. Results from mixed effects regression analysis on data from the 15 patients identified platelet count (and thrombocytopenia) and urea as significant independent predictors of concordance between APR and N-ACT.

We report a moderate degree of positive correlation between APR and N-ACT. We conclude that there is poor agreement between the ACT and aPTT for the heparin concentrations in patients supported with ECMO. Our results indicate that platelet count and urea are significant independent variables affecting concordance between ACT and aPTT measurements.