Partial thromboplastin time is more predictive of bleeding than anti-Xa levels in heparinized pediatric patients after cardiac surgery.

Abstract

Anticoagulation with unfractionated heparin (UFH) after pediatric cardiac surgery can be monitored using either activated partial thromboplastin time (aPTT) or anti-factor Xa activity (anti-Xa). However, correlation of bleeding with either of these laboratory values has not been established. We sought to determine the correlation between bleeding events and aPTT and anti-Xa in patients who undergo anticoagulation after congenital heart surgery. We prospectively studied pediatric patients treated with UFH after cardiac surgery over an 11-month period. Bleeding events were prospectively assessed and adjudicated. The highest aPTT and corresponding anti-Xa for the 24 hours before bleeding events were collected to assess for association with bleeding. Statistical analysis was performed using generalized additive logistic regression. A total of 202 patients received UFH over 1488 patient-days. The median age at surgery was 0.4 years (interquartile range, 0.1-2.2). A total of 45 major or clinically relevant bleeding events were observed. The correlation between aPTT and anti-Xa was of moderate strength (R = 0.58; P < .001). The odds of bleeding increased significantly when aPTT exceeded 150 (odds ratio, 1.71 per 10-second increase in aPTT, 95% confidence interval, 1.21-2.42; P = .003). Anti-Xa was not associated with bleeding (odds ratio, 1.11 per 0.1 IU/mL increase, 95% confidence interval, 0.89-1.29; P = .34). In heparinized pediatric patients after cardiac surgery, increased risk of bleeding is more closely associated with elevated aPTT levels than elevated anti-Xa levels. In addition to anti-Xa, monitoring of aPTT levels should be considered during titration of UFH in pediatric patients after cardiac surgery.