Individualized anticoagulation during cardiopulmonary bypass (CPB) may reduce drug exposure without sacrificing control of the circuit. In a prospective randomized trial of 100 adults undergoing isolated coronary artery bypass grafting (CABG), an activated clotting time (ACT)-guided strategy was associated with lower 24-hour chest drain output and fewer transfusions than conventional weight-based heparin dosing. The findings are promising, but the small, unregistered single-center trial does not establish thrombotic safety or a universal dosing protocol.
Why individualize heparin and protamine?
Weight-based heparin dosing provides a practical starting point, but patients vary in their anticoagulant response. Reversal is equally important: giving protamine according to the total heparin administered may not accurately reflect the amount remaining at the end of bypass. El Dsouki and colleagues investigated whether an ACT-based calculation for both agents could improve postoperative coagulation and reduce bleeding in patients undergoing isolated CABG.
The study addresses an everyday perfusion question: can a readily available monitoring measurement support more selective anticoagulant and reversal exposure? Its results should be read as evidence about one defined protocol and population, rather than proof that all cardiac surgical patients should receive less heparin or protamine.
Trial design and bypass conditions
The investigators randomized 100 adults equally between ACT-guided management and standard care. Recruitment occurred from December 2024 through July 2025. A computer-generated block sequence and sealed opaque envelopes were used for allocation. Anesthesia and perfusion teams could not be blinded, but postoperative outcome assessment and statistical analysis were blinded. All randomized patients were included in the final intention-to-treat analysis, with no reported losses to follow-up.
Patients underwent elective or urgent isolated CABG. Emergency surgery, significant uncorrected coagulopathy, inadequately discontinued antithrombotic treatment and low baseline antithrombin activity were among the exclusions. The circuit used a roller pump and membrane oxygenator with coated tubing. Both groups received normothermic perfusion at 36–37°C and the same goal-directed perfusion approach. These conditions matter when considering whether results apply to different circuits, temperature strategies or operations.
What the ACT-guided strategy involved
The intervention calculated initial and supplemental heparin from the difference between the measured ACT and a target of at least 400 seconds, using an assumed response of approximately 1.3 seconds per IU/kg. Protamine was calculated from an ACT-based estimate of residual heparin after bypass. The comparison group received an initial heparin bolus of 300 IU/kg and conventional additional dosing and empirical reversal.
The assumed heparin response was an operational estimate, not a fixed biological constant. ACT is also influenced by factors beyond heparin concentration, so an ACT-derived estimate of residual heparin should not be treated as a direct drug assay. The study target likewise does not establish a universal safe ACT threshold across devices and perfusion settings.
Rotational thromboelastometry (ROTEM) was assessed before, during and after surgery. It supported postoperative hemostatic management and transfusion decisions when indicated, but did not directly determine intraoperative heparin or protamine doses. The trial therefore evaluated a combined perioperative strategy rather than isolating every contribution of dosing, monitoring and transfusion practice.
Less bleeding and lower transfusion exposure
The primary endpoint was chest drain output during the first 24 postoperative hours. Mean output was 560 ± 210 mL with ACT-guided management and 820 ± 310 mL with standard care (P < 0.001). The authors reported an absolute mean reduction of 260 mL, with a 95% confidence interval of 160–360 mL.
Any blood product transfusion occurred in 14 of 50 patients (28%) in the guided group and 26 of 50 (52%) in the standard group (P = 0.01), an absolute difference of 24 percentage points. Table 6 reports red blood cell use of 0.6 ± 1.0 versus 1.4 ± 1.6 units (P = 0.003). This distinction is relevant because the abstract describes these same unit values as total allogeneic units, whereas the table labels them as RBC transfusion.
Initial heparin exposure was 214 ± 22 versus 300 IU/kg, and total exposure was 248 ± 34 versus 332 ± 41 IU/kg. Total protamine was 178 ± 32 versus 246 ± 45 mg; all three comparisons had P < 0.001. Additional protamine was required in 12% versus 36% of patients. The authors also reported fewer supplemental heparin boluses and a higher proportion of ACT measurements within the study target range.
Coagulation findings and clinical outcomes
After reversal, ROTEM findings favored the guided strategy: EXTEM clotting time was 71 ± 14 versus 89 ± 21 seconds, and EXTEM maximum clot firmness was 61 ± 7 versus 55 ± 8 mm. FIBTEM maximum clot firmness was also higher. These measurements support a difference in postoperative coagulation profiles, but they are not substitutes for definitive clinical safety endpoints.
Platelet count reduction was smaller, at 22% versus 31%. ICU and hospital stays were shorter in the intervention group, but these were secondary outcomes and should be considered exploratory. Re-exploration for bleeding occurred in one versus four patients and did not differ significantly. There were no thrombotic events in the guided group and one in the comparison group; no patient died within 30 days.
Those small event counts cannot demonstrate equal thrombotic safety or exclude uncommon harm. The sample size calculation addressed bleeding, not thrombosis, mortality or the other secondary outcomes. A lack of statistical significance for a rare event is not evidence of equivalence.
Limitations and reporting questions
The single-center setting, 100-patient sample and restriction to isolated CABG limit generalizability. The investigators state that the trial was not prospectively registered in a public registry. Together with open-label treatment, this reduces confidence in the prespecification and reproducibility of the study. ROTEM availability may also have influenced transfusion decisions, and the trial did not use a fully standardized goal-directed transfusion algorithm.
Some dosing summaries warrant clarification. Table 4’s baseline-ACT strata imply an average initial dose of approximately 209 IU/kg, while Table 3 reports 214 IU/kg. The stated formula also implies little initial-dose variation across the baseline ACT range shown in Table 4, compared with the reported standard deviation of 22 IU/kg. These observations do not invalidate the bleeding endpoint, but they make independent validation of the dosing algorithm and source data important.
Several secondary endpoints were tested, and the authors identify them as exploratory. A formal cost analysis was not performed, so lower drug and transfusion exposure should not be presented as proven cost-effectiveness. Larger, prospectively registered multicenter trials are needed to assess reproducibility, algorithm calibration and uncommon adverse events.
Relevance for perfusion practice
This trial adds evidence that individualized heparin and protamine exposure deserves further study within an integrated coagulation strategy. For perfusion teams, its strongest message is the importance of evaluating anticoagulation, reversal, bleeding and transfusion together. It supports critical review of local data and protocols, while leaving important questions about safety and wider applicability unresolved.
Source: El Dsouki Y, Condello I, Gunertem OE, Yildirim A, Gunaydin S, Lorusso R. Optimizing Heparin and Protamine Dosing During Cardiopulmonary Bypass Using Activated Clotting Time-Guided Protocols: A Single-Centre Prospective Randomized Controlled Trial. Interdisciplinary CardioVascular and Thoracic Surgery. 2026;41(8):ivag197. doi:10.1093/icvts/ivag197.
Funding and disclosures: No funding was declared. Two authors report employment with Spectrum Medical; the authors state that the company provided no support or influence. Additional industry relationships are disclosed in the original paper. The source article is open access under CC BY-NC 4.0.
