Introduction: Neonates are especially susceptible to bleeding during and after cardiopulmonary bypass (CPB) surgery seemingly by virtue of a proportionately larger foreign surface exposure area for blood, significant hemodilution, and a coagulation system that differs from older patients. During CPB, heparin is administered to prevent thrombin formation. After CPB, it is reversed with protamine sulfate. Although administered to reverse anticoagulation, high levels of protamine sulfate paradoxically inhibit blood coagulation. This study evaluated the effect of heparin anticoagulation and protamine sulfate reversal on the accuracy of diagnostic coagulation tests and bleeding outcomes in a neonatal population undergoing CPB for congenital heart defects in the first 30 days of life.

Methods: Blood samples were obtained at baseline (Draw 1), on CPB (Draw 2), following protamine reversal (Draw 3) and upon admission to the cardiac intensive care unit (Draw 4). Thrombin time (±protamine), FXa activity, activated partial thromboplastin time (APTT), Rotational Thromboelastometry (ROTEM) and Calibrated Automated Thrombography (CAT) were performed. Excessive bleeding was defined by meeting one or more of the following criteria: ≥ 84 mL/kg total chest tube output (CTO) in the first 24 hours, ≥ 7 mL/kg/hr CTO for ≥2 consecutive hours in the first 12 hours, or re-exploration in the first 24 hours for bleeding or tamponade.

Results: This prospective observational study enrolled 44 neonates. Of these, 16 (36%) had excessive bleeding. The thrombin time assay was sensitive to heparin but was only minimally impacted by protamine sulfate, unless heparin was in the sample. Therefore, it was selected as the "gold-standard" assay for measurement of heparin. Relative heparin levels, determined by rate of fibrin formation, inversely correlated with thrombin generation measured using CAT, particularly in comparison to the thrombin generation rate in Draw 4 samples (exponential fit R2= 0.333). Higher plasma heparin concentration was present in 85% of the patients at Draw 4 than at Draw 3, indicating postoperative heparin rebound. This was important because, although the plasma heparin concentration in bleeders varied greatly, those patients with the two highest heparin concentrations at Draw 4 had excessive bleeding. The FXa activity assay was altered by protamine and by heparin. Therefore, samples with elevated FXa activity but normal thrombin time likely had high protamine. There were 4 such samples and 2 were bleeders. Additional coagulation assays were compared to the thrombin time assay to assess their performance for detection of heparin. ROTEM and APTT assays also correlated weakly (R2= 0.0531 with P value 0.1527 and R2 = 0.1549 with P value 0.0099, respectively) with the thrombin time assay at Draw 4. However, no correlation was present at Draw 3.

Conclusions: Bleeding remains a common complication in neonates undergoing CPB for repair of heart defects. While there are many possible causes for the bleeding, excess heparin or protamine may contribute to its severity in a subset of cases. The simple to perform and inexpensive thrombin time test readily and accurately detects excess heparin. Increased use of this test may be beneficial, particularly for detection of heparin rebound that is observed in approximately 85% of patients. Excess protamine may also contribute to bleeding in these patients, and can be assessed using a combination of thrombin time and factor Xa activity assays.


Mast: Novo Nordisk: Research Funding.

Author notes


Asterisk with author names denotes non-ASH members.