Factors Associated With Discordance Between the Activated Partial Thromboplastin Time and Anti-factor Xa Activity During Extracorporeal Membrane Oxygenation: A Retrospective Cohort Study.

BACKGROUND: Reliable anticoagulation monitoring during extracorporeal membrane oxygenation (ECMO) guides the balance between bleeding and thrombosis, yet the activated partial thromboplastin time (aPTT) and anti-factor Xa (anti-Xa) activity frequently disagree, and the factors associated with discordance are poorly defined. We sought to determine the prevalence of aPTT and anti-Xa discordance during unfractionated heparin anticoagulation on ECMO, to identify associated factors, and to characterize clinician response. METHODOLOGY: We conducted a retrospective cohort study at a single ECMO center from 2019 to 2025, including neonates, children, and adults anticoagulated with heparin. aPTT and anti-Xa values drawn within six hours of each other were paired and classified as concordant or discordant. Generalized linear mixed-effects models with a random intercept for patient accounted for repeated measurements within patients. RESULTS: A total of 89 ECMO runs contributed 1,252 paired observations. Discordance occurred in 59.1% (740 of 1,252) of pairs, and the correlation between assays was moderate (Pearson R=0.41). Discordance was most frequent in neonates (464 of 671, 69.2%) compared with older children (32 of 66, 48.5%) and adults (244 of 515, 47.4%; unadjusted p<0.001). In univariate analysis, discordance was associated with lower weight and lower fibrinogen and with higher anti-Xa, aPTT, prothrombin time, and heparin dose. In mixed-effects models, a higher heparin infusion dose at the time of paired sampling was the only factor independently associated with discordance (adjusted odds ratio: 1.14; 95% CI: 1.07-1.22; p<0.001). Clinicians adjusted heparin after 396 of 740 discordant pairs (53.5%); changes aligned with anti-Xa in 329 of 740 (44.5%), with aPTT in 238 of 740 (32.2%), and with neither assay in 173 of 740 (23.4%). Anti-Xa-aligned changes were associated with a higher baseline heparin dose (adjusted odds ratio: 1.10; 95% CI: 1.02-1.19), whereas aPTT-aligned changes were associated with a higher aPTT (adjusted odds ratio: 1.04; 95% CI: 1.01-1.06) and higher fibrinogen (adjusted odds ratio: 1.01; 95% CI: 1.00-1.01). CONCLUSIONS: Discordance between aPTT and anti-Xa is common during ECMO and varies by age group. Higher heparin infusion dose was independently associated with discordance, and clinician responses to discordant pairs were heterogeneous, highlighting persistent uncertainty in managing discordant anticoagulation assays.