Does the initial heparin dose influence anticoagulation during neonatal cardiopulmonary bypass (CPB), even when subsequent doses are guided by activated clotting time (ACT)? A randomized trial from Fuwai Hospital compared initial systemic doses of 600 and 400 U/kg in 76 neonates with congenital heart disease. The higher initial dose produced a smaller increase in a marker of thrombin generation and more frequent achievement of the first ACT target.
The findings provide useful mechanistic evidence, but they do not establish a universal neonatal heparin dose or demonstrate a definitive clinical benefit. Both groups received ACT-guided supplementation, and the trial was powered for a laboratory endpoint rather than bleeding, transfusion, or mortality.
Study Design and Anticoagulation Protocol
Wang and colleagues conducted this prospective, single-center trial between February 2023 and December 2024. Seventy-six patients aged 28 days or younger were randomized equally to the two initial-dose strategies. The study excluded patients with specified preoperative coagulation or major organ disorders and those already receiving extracorporeal life support.
In both groups, the study protocol required ACT >480 seconds before starting and during CPB. When the first dose did not achieve that target, additional heparin was administered in 200 U/kg increments, with repeat ACT testing. These doses and thresholds describe the research protocol; they are not a recommendation to change an institution’s neonatal anticoagulation policy.
Allocation was concealed until assignment. Surgeons, ICU physicians, outcome assessors, laboratory personnel, and the patients’ families were blinded. Perfusionists and anesthesiologists administering anticoagulation were not blinded.
What the Primary Endpoint Measured
The primary endpoint was the change in prothrombin fragment 1+2 (F1+2), a marker of prothrombin activation and thrombin generation. The first sample was obtained within five minutes after the assigned initial heparin dose, before supplemental heparin and before CPB. The second was obtained at CPB termination, before protamine.
This timing matters: “pre-CPB” did not mean a pre-heparin baseline. The endpoint compares the change from the early post-dose state to the end of bypass. It provides complementary information to ACT, rather than replacing ACT as the procedural monitoring test.
Key Results
| Measure | 600 U/kg initial dose | 400 U/kg initial dose |
|---|---|---|
| Randomized patients | 38 | 38 |
| Mean change in F1+2 | 0.70 nmol/L | 4.08 nmol/L |
| ACT >480 seconds after initial dose | 84.2% | 44.7% |
| Median surgical preparation time | 31.50 minutes | 36.50 minutes |
| Median total systemic heparin | 663.62 U/kg | 640.87 U/kg |
| Median mechanical ventilation duration | 45.00 hours | 52.50 hours |
| In-hospital deaths | 1/38 | 0/38 |
Selected results as reported. Total systemic heparin excludes the 1,000 U added to the circuit prime. Secondary clinical comparisons were exploratory and were not adjusted for multiple testing.
Less thrombin generation and a stronger first ACT response
The treatment-minus-control difference in mean F1+2 change was −3.38 nmol/L (95% CI −4.58 to −2.18; P<0.001). Initial ACT target attainment was 39.5 percentage points higher with 600 U/kg; the reported risk ratio was 1.88 (95% CI 1.29–2.75). Complete-case and other sensitivity analyses supported the primary finding.
Initial dose differed, but total systemic dose did not significantly differ
The higher initial-dose group required less supplemental heparin before and during CPB. However, the total systemic dose across the procedure did not differ significantly (P=0.723). This is therefore a comparison of initial dosing strategies followed by responsive supplementation, rather than a simple comparison of two fixed total exposures.
Surgical preparation time, defined as skin incision to CPB initiation, was shorter in the 600 U/kg group. Reduced repeat dosing and testing may help explain that finding, but the unblinded clinicians and the broad definition of preparation time limit a firm causal interpretation of the workflow effect.
Bleeding, Transfusion, and Safety
Exploratory analyses found lower 72-hour chest drainage and albumin use and shorter mechanical ventilation duration with the higher initial dose. Red-cell, plasma, and platelet transfusion requirements did not differ significantly, and most other clinical outcomes were not significantly different.
One in-hospital death occurred in the 600 U/kg group and none in the 400 U/kg group. Major bleeding occurred in none and one patient, respectively. These very small event counts cannot establish equivalent safety or rule out important benefits or harms. A nonsignificant result in a small trial is not proof that the strategies have the same clinical effects.
Relevance to Perfusion Practice
The study highlights a distinction between reaching an ACT threshold and limiting biochemical coagulation activation throughout bypass. In the studied population, changing the initial dose improved first-pass ACT response and reduced the rise in F1+2 even though total systemic heparin exposure was similar.
For perfusion teams reviewing neonatal protocols, the results support discussion of initial dosing, response variability, the timing of repeat testing, and how supplemental doses are documented. They do not justify adopting 600 U/kg for every neonate. Interpretation must remain tied to the center’s circuit, monitoring methods, patient population, and ACT-guided supplementation protocol.
Limitations and Evidence Assessment
Randomization, allocation concealment, blinded laboratory assessment, and sensitivity analyses strengthen the mechanistic result. Nevertheless, this was a small, single-center trial centered on a surrogate endpoint. Clinicians managing anticoagulation were unblinded, follow-up was limited to the early postoperative period, and secondary outcomes were not corrected for multiple comparisons.
Two patients, one in each group, required ECMO within 24 hours and lacked paired F1+2 measurements. They remained in the primary intention-to-treat analysis through multiple imputation; complete-case analyses included 74 patients. The missingness is clinically relevant, so the imputation assumptions and sensitivity analyses deserve attention.
The ACT analyzer’s ceiling of 999 seconds also limited measurement of very prolonged values. Most importantly, the sample size was insufficient to establish effects on uncommon adverse events or patient-centered outcomes. Larger multicenter trials powered for clinical benefit and safety are needed before treating this as practice-changing evidence.
Source and Attribution
This original IPA educational summary is based on Wang and colleagues’ open-access trial, published under the Creative Commons Attribution–NonCommercial 4.0 license. Findings and selected numerical results have been summarized and their presentation adapted. The cover is a conceptual illustration, not an anatomical or circuit-assembly guide.
Original Article
Wang H, Liu Q, Gao P, et al. Optimization of Anticoagulation Strategy During Cardiopulmonary Bypass in Neonates with Congenital Heart Disease: A Randomized Controlled Trial. Drug Design, Development and Therapy. 2026;20:634756. Published October 15, 2026. doi:10.2147/DDDT.S634756. Trial registration: ChiCTR2300069157.
