Comparison of aprotinin and tranexamic acid in the prevention of perioperative bleeding in non-elective cardiac procedures involving the ascending aorta

Introduction

Despite substantial advancements in perioperative management, bleeding remains a major complication in patients undergoing non-elective surgery involving the ascending aorta. These procedures usually combine complex surgical repair with a frequent use of hypothermia, prolonged cardiopulmonary bypass (CPB) duration, and specific disruptions in hemostatic physiology, making them particularly prone to significant blood loss. Historically, prevention of excessive fibrinolysis using aprotinin—a broad-spectrum protease inhibitor—served as a mainstay to reduce perioperative bleeding in cardiac surgery (1). This drug was suspended in 2007 following the Blood Conservation Using Antifibrinolytics in a Randomized Trial (BART) study (2), which called into question its safety of use, notably because of an increased risk of mortality compared to lysine analogues such as tranexamic acid. Aprotinin was also associated with an increased risk of acute kidney failure. Since this publication and the aprotinin withdrawal, tranexamic acid was the main antifibrinolytic agent used in the perioperative period. It is known for its ability to reduce the need for transfusions, with satisfactory safety (3,4). Its use for type A aortic dissection was also associated with a reduction in bleeding (5).

Patients included in the BART study were mostly patients undergoing coronary artery bypass grafting, which is difficult to transpose to emergent aortic surgery. Recent observational studies and meta-analyses revisited the safety profile of aprotinin in selected high-risk patients, suggesting that it may confer bleeding reduction and potential survival benefit, particularly when compared with lysine analogs such as tranexamic acid (3). Indeed, among high-risk patients aortic surgeries present the highest risk of moderate or severe bleeding (6), suggesting that aprotinin may be of interest in this specific population. For example, 16.7% of patients included in the Nordic aprotinin patient registry (NAPaR) presented with aortic dissection (7). Because evidence comparing aprotinin and tranexamic acid in non-elective cardiac surgery involving the ascending aorta remains limited and heterogeneous, we aimed to provide additional real-world data regarding the efficacy and safety of these two antifibrinolytic strategies in this high-risk bleeding population.

The objective of our retrospective monocentric study was to compare the efficacy and safety of aprotinin and tranexamic acid in patients operated on the ascending aorta for emergent acute aortic syndrome. We present this article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-0931/rc).

Methods

Ethics

This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. According to French regulations for retrospective observational studies using anonymized routine-care data, formal approval from an Institutional Review Board or Ethics Committee and written informed consent were not required. The data collection adhered to the MR004 methodology from the national committee for informatics and freedom (CNIL).

Study design

We conducted a monocentric retrospective observational cohort study in a tertiary hospital. Because the use of aprotinin was reintroduced in our center in 2020, two different periods of time were compared. The first cohort included patients receiving tranexamic acid from January 2016 to July 2018, whose data were extracted from a previously published cohort (6). The second cohort included patients receiving aprotinin from November 2020 to March 2022.

Study population

The inclusion criterion was age ≥18 years with cardiac surgery with CPB and including the ascending aorta in a non-elective setting.

Exclusion criteria were other surgical procedures, previous inherited coagulation disorders and patients <18 years old.

Data collection

Clinical, demographic, and procedural data were extracted from our electronic medical records: CDP2, M-EVA, IntelliSpace Critical Care & Anesthesia (ICCA) Adult, and anesthesia charts. Different variables were collected, including age, sex, body mass index (BMI), comorbidities, preoperative laboratory blood tests (platelet count, fibrinogen concentration, hemoglobin concentration, creatinine), CPB and aortic clamping durations and the presence of ongoing antithrombotic or anticoagulant therapy. The European System for Cardiac Operative Risk Evaluation II (EuroSCORE II) (8), TRUST (Transfusion Risk Understanding Scoring Tool) score (9) and the PeriOperative Bleeding Score in Cardiac Surgery (POBS-Card) (6) were recorded to characterize the baseline risk.

Surgery

Cardiac surgery and CPB followed the current guidelines from the European Association for Cardio-Thoracic Surgery (EACTS) (10). A drainage cannula was placed in the right atrium (except for mitral valve surgery) thanks to a double-lumen cannula and the arterial cannulation was performed through the right axillary artery and/or the femoral artery according to the surgeon and the case. Surgery requiring a distal anastomosis under circulatory arrest benefited from central hypothermia (28–31 °C), and left common carotid artery may have been cannulated by a Gundry cannula if necessary; the right common carotid artery was perfused either via the reperfusion cannula in the right axillary artery (if present, with clamping at the base of the brachiocephalic artery), or via a Gundry cannula. Carotids were perfused with blood at 15 °C during circulatory arrest.

Cardioplegia was done using cold blood mixed with a histidine, tryptophan, ketoglutarate-enriched fluid (Perisoc^®, Franz Khöler Chemie GmbH, Germany).

CPB was ensured by a Stockert S5 rotational pump at a pump flow of at least 2.4 L/min/m², in combination with an Inspire 6 or 8 PHISIO single chamber circuit (LivaNova, France) or a CAPIOX F15 (Terumo, France), according to availability and patient characteristics. Pump flow was adjusted to obtain an arterial delivery of oxygen of at least 280 mL/min/m², and if necessary red blood cells may have been administered to reach a minimum hematocrit of at least 20%. The administration of these packed red blood cells was not part of the universal definition of perioperative bleeding (UDPB) classification.

Anticoagulation and neutralization

Heparin was administered before CPB at the dose of 10,000 IU/m² of body surface area, in order to achieve an activated clotting time (ACT) of at least 400 sec (Hemochron Elite Signature, Werfen, France), and additional doses may have been administered to maintain ACT above this aim. Neutralization of heparin was performed after CPB weaning using protamine at 1:1 of the dose of heparin for every dose administered within 60 min, and 0.6:1 for every dose >60 min.

Intraoperative bleeding was assessed 10 min after neutralization of heparin. The surgeon performed a checklist of potential sites of bleeding as previously described (11). In cases of non-surgical bleeding, coagulation disorders were handled according to an institutional protocol including the use of viscoelastic testing (Quantra^®, Stago, or TEG-based systems, Haemonetics, France), presented in Figure S1. In cases of unavailability of viscoelastic testing, fibrinogen concentrates were administered if plasma fibrinogen <2 g/L, platelet transfusion if platelet count <100 G/L (or if active antiaggregant therapy or CPB duration >120 min) and fresh frozen plasma if prothrombin ratio <50 %. Red blood packs were administered if hemoglobin <8 g/dL off-CPB, or if hematocrit <25% on-CPB, or in cases of uncontrolled hemorrhagic shock.

Antifibrinolytics administration

After induction of anesthesia, patients of historical cohort received an intravenous bolus of tranexamic acid (doses between 1 and 2 g as a single bolus). The second cohort comprised patients admitted at Rouen Hospital and receiving aprotinin from November 2020 to March 2022 whose data were extracted from the local dataset of the Nordic Aprotinin Patient Registry (NAPaR). The protocols used for aprotinin administration were those recommended by the ARCOTHOVA society (“Anesthésie Réanimation Coeur Thorax Vaisseaux”, translated as “Anesthesia Reanimation Heart Thorax Vessels”). The protocol for half-dosage aprotinin was 1×10⁶ KIU during the induction of anesthesia, 1×10⁶ KIU in the CPB priming, then a continuous infusion at 250,000 KIU/h. The protocol for full-dosage aprotinin was 2×10⁶ KIU during the induction of anesthesia, 2×10⁶ KIU in the CPB priming, then a continuous infusion at a rate of 500,000 KIU/h.

Outcomes

The primary endpoint was the bleeding severity, quantified by the UDPB score as published by Dyke et al. (12). This validated scoring system incorporates chest tube drainage, transfusion requirements, re-exploration for bleeding and other clinical parameters into a five-level ordinal scale from “no bleeding or trivial bleeding” (UDPB =0) to “massive or life-threatening bleeding” (UDPB =4). In this study, this score was dichotomized into two categories that have clinical relevance: severe or massive bleeding (UDPB score 3–4) versus insignificant to moderate bleeding (UDPB score 0–2). Secondary outcomes encompassed the individual components of the UDPB score (including blood products usage and the need for re-exploration within the first postoperative 24 hours), acute kidney injury (AKI) according to the “Kidney Disease Improving Global Outcomes” (KDIGO) classification, ischemic stroke, myocardial infarction, infection, in-hospital length of stay, and 30-day mortality. For this study, the KDIGO classification was based on the worst creatinine level within the first 7 postoperative days only, without taking into account diuresis because of the regular use of diuretics.

Statistical analysis

Continuous variables were expressed as mean with standard deviation or median with interquartile ranges according to the distribution of data, as explored using a Shapiro-Wilk test. Categorical variables were expressed by their frequency and percentage. Comparisons were performed using chi-squared test, Mann-Whitney test, or Fisher’s exact test, as appropriate.

The association of aprotinin use with the occurrence of bleeding was explored using a multivariable analysis using a binary logistic regression model including the following variables: age, sex, POBS-Card, level of emergency and CPB duration. For this analysis, deviance was 61.8, the Akaike information criterion (AIC) was 75.8, R²_MCF was 0.215. All the variance inflation factors (VIFs) were <1.5. According to the cross-validation test, accuracy was 76.38%±3.76% and kappa at 0.19±0.13.

A P value <0.05 was considered statistically significant for all the analyses. Statistics were performed using DATAtab, Excel, XLSTAT, jamovi and R. No sensitivity analyses were performed, as the model was validated using statistical quality measures (AIC, R², VIF, cross-validation). Missing data were not imputed and analyses were performed on available data only.

Results

Participants and descriptive data

A total of 72 ascending aorta procedures met the inclusion criteria: 32 from January 2016 to July 2018 concerning the tranexamic acid group and 40 from November 2020 to March 2022 concerning the aprotinin group (Figure 1).

Figure 1 Flow chart. CABG, coronary artery bypass graft; CPB, cardiopulmonary bypass; DAPT, dual antiplatelet therapy; LVAD, left ventricular assist device.

The distribution of surgical indications differed significantly between groups (Table 1; P=0.008). Patients were not significantly different in age, morphological criteria and preoperative risk scores (Table 1). Preoperative fibrinogen plasma level and platelet count were significantly lower in the aprotinin group whereas the prothrombin ratio was significantly higher. The duration of aortic clamping was not significantly different. There were more patients operated with use of circulatory arrest in the aprotinin group (80% vs. 53%, P=0.02), and the duration of circulatory arrest was significantly longer in this group [24 (Jeny15; 35) vs. 12 (0; 28) min, P=0.02].

Primary outcome

Among the 40 patients who received aprotinin, 4 received the full-dose regimen and 36 received a half-dose.

No difference was observed between groups for major bleeding defined by a UDPB score ≥3, with 10 patients (31.25%) in the tranexamic acid group and 7 patients (17.5%) in the aprotinin group (P=0.26).

Results for each UDPB class are presented in Figure 2. No significant difference was observed in the overall distribution of UDPB categories between groups (chi-squared test, P=0.34).

Figure 2 Frequency of UDPB score categories in patients treated by tranexamic acid or aprotinin. Chi-squared test was used to compare each category of UDPB score (P=0.34). UDPB score, Universal Definition for Perioperative Bleeding score.

Multivariable analysis after adjustment for gender, age, emergency level, POBS-Card and duration of CPB showed that aprotinin was associated with a significant reduction in bleeding (odds ratio: 0.19, 95% confidence interval: 0.05–0.79, P=0.02), whereas emergent surgery and longer CPB duration were significantly associated with higher bleeding (Table 2).

Secondary outcomes

Among the UDPB score criteria, the administration of PCCs (50% vs. 25%, P=0.03) and the need for fibrinogen (77.5% vs. 53.13%, P=0.02) were significantly higher in the aprotinin group, whereas the administration of rFVIIa (2.5% vs. 18.8%, P=0.04) was more frequent in the tranexamic acid group (Table 3).

Neither the worst serum creatinine level during the first 7 postoperative days, nor the need for renal replacement therapy, nor the KDIGO score were significantly different between the two groups.

After the intervention and before either discharge or re-intervention with CPB, no significant difference was found between the two groups in terms of deep vein thrombosis, pulmonary embolism, ischemic stroke, myocardial infarction and revision surgery for bleeding.

No difference was observed for length of intensive care unit and hospital stays or 30-day mortality.

Discussion

Tranexamic acid is a synthetic lysine analogue that binds to the lysine-binding sites of plasminogen. This binding prevents the interaction between plasminogen and lysine residues on fibrin, thereby reducing its activation into plasmin at the fibrin clot surface and limiting fibrin clot degradation. Aprotinin is a non-specific serine protease inhibitor of polypeptide origin. It inhibits several enzymes involved in fibrinolysis and inflammation, particularly plasmin and kallikrein. By directly inhibiting plasmin at its catalytic site, aprotinin reduces fibrin degradation and limits fibrinolysis. Kallikrein inhibition also contributes to reducing inflammatory activation and coagulation factor consumption associated with CPB.

Recent patient blood management guidelines continue to recommend antifibrinolytic agents in high-risk cardiac surgery to reduce perioperative bleeding and transfusion requirements (13). Aprotinin is a highly controversial drug with either beneficial or harmful effects according to studies and populations studied.

One possible explanation for the observed association between aprotinin and reduced bleeding in our study may relate to the specific pathophysiology of emergent ascending aortic surgery. These procedures frequently combine prolonged CPB, hypothermia, circulatory arrest, and extensive activation of the contact coagulation pathway, resulting in profound hemostatic disturbances and hyperfibrinolysis. Beyond its antifibrinolytic properties, aprotinin exerts broad serine protease inhibition, including effects on plasmin and kallikrein pathways, which may theoretically provide additional protection against perioperative coagulopathy in this specific setting.

We showed here a significant association between aprotinin and reduction in major bleeding after adjustment for several variables of interest. Among them, POBS-Card score has been described as providing an acceptable prediction of the risk of bleeding, allowing to optimize adjustment and limiting selection bias in such a heterogeneous population (6). This score takes into account BMI, redo surgery, prothrombin rate, APTTr, thrombocytopenia, fibrinogen level, combined surgery, aortic surgery, SAPT and DAPT. Gallo et al. published a multicentric retrospective study, in which 693 patients were analyzed, with all the levels of bleeding risk (14). No difference was found in the occurrence of UDPB score 3 or 4 between half-dose aprotinin and moderate-dose tranexamic acid. This study included 147 redo surgeries; 50 of those involved the ascending aorta. There were significantly more thromboembolic events in the aprotinin-treated group, but no significant difference in 30-day mortality, need for surgical re-exploration, need for transfusions (except on unadjusted analysis), ICU length of stay and hospital length of stay. A limitation of this study was that intraoperative ACT was significantly lower in the tranexamic acid group. Because no subgroup analysis of aortic surgeries was performed, the results are not transposable to our study.

In our study, the inclusion period for patients undergoing emergency ascending aortic surgery treated with tranexamic acid was approximately two and a half years, compared with one and a half years for aprotinin. Nevertheless, we included more patients in the aprotinin arm. We cannot explain this increase in the number of procedures for this indication; however, it does demonstrate that in our center, the overwhelming majority of patients undergoing emergency ascending aortic surgery are preferentially treated with aprotinin rather than tranexamic acid when both treatments are available, so using the same period for both cohorts would cause an obvious bias by indication.

Another limitation concerns the tranexamic acid regimen used in the historical cohort. During the inclusion period (2016–2018), tranexamic acid was administered as a fixed intravenous bolus of 1–2 g without continuous infusion, reflecting routine clinical practice in our institution at that time. Higher-dose weight-adjusted regimens remained controversial because of concerns regarding thrombotic complications and postoperative seizures. More standardized protocols using weight-adjusted dosing and continuous infusion progressively emerged later, notably following the publication of the OPTIMAL trial (15) and subsequent meta-analyses in high-risk cardiac surgery. Consequently, it cannot be excluded that the tranexamic acid protocol used in our study underestimated its efficacy compared with contemporary dosing strategies.

Although our sample size does not allow us to demonstrate a statistically significant difference in 30-day mortality, the mortality rate was more than five times higher in the aprotinin group than in the tranexamic acid group. It is important to emphasize this point in order to maintain a certain degree of caution when interpreting our results. Nonetheless, several factors may contribute to this difference. First, patients in the aprotinin group presented a more severe perioperative profile, with significantly more acute aortic syndromes, longer circulatory arrest durations, and longer circulatory arrest durations. Furthermore, because the two cohorts originated from different time periods, unmeasured differences in perioperative management and transfusion practices may also have contributed to the observed mortality difference. Moreover, there were more immediate emergency procedures in the aprotinin group than in the tranexamic acid group, with a difference close to statistical significance.

Patients in the aprotinin group underwent circulatory arrest more frequently and for longer durations, reflecting the greater procedural complexity of this cohort. Although circulatory arrest duration was not included separately in the multivariable model, CPB duration was retained because it is closely related to circulatory arrest, hypothermia, and overall operative complexity. Including both variables simultaneously in a model with a limited number of bleeding events would likely have increased collinearity and the risk of overfitting.

The more frequent use of fibrinogen and PCCs in the aprotinin group may be due to changes in hemostasis management practices over time or the preferential use of aprotinin in the worst patients (because tranexamic acid was no longer available). The persistence of the association between aprotinin use and lower bleeding severity despite these potential sources of bias remains an interesting finding, although causality cannot be established in this retrospective study. Because of the retrospective design and the inclusion of two historical cohorts from different time periods (January 2016–July 2018 for tranexamic acid and November 2020–March 2022 for aprotinin), we agree that changes in transfusion and hemostatic management practices over time may represent a source of bias. Although viscoelastic testing was already used during both study periods through TEG-based monitoring, the evolution of coagulation management practices during the aprotinin period may have contributed to a more aggressive correction of coagulation abnormalities, independently of the antifibrinolytic agent used.

Reidy et al. published a retrospective study of 250 patients operated for type A aortic dissection and found no difference in terms of transfusions, surgical revision for bleeding within the first 12 hours after surgery, delayed sternal closure, AKI, stroke or death between patients treated with aprotinin and those treated with tranexamic acid (16). Similarly, in a retrospective single-center study of 78 patients treated for acute aortic syndrome, Laborier et al. (A) obtained similar results in terms of transfusions and bleeding in the first 24 hours after surgery, thromboembolic events, need for extrarenal purification and death, but still found a signal of higher renal risk for aprotinin compared with tranexamic acid.

The limited number of patients included in our study did not allow us to explore differences between full-dose and half-dose aprotinin treatment. A single-center observational study of 8,548 patients found no dose effect of aprotinin on renal dysfunction, the latter being correlated instead with preoperative renal dysfunction and the complexity of the surgical procedure (17). The randomized trial by Levy et al. (18) comparing full-dose and half-dose aprotinin found no difference in bleeding through the drains. However, this was a small study: 15 patients were treated with full-dose and 9 with half-dose. Conversely, a recent post-hoc analysis of the NAPaR showed that full-dose was associated with less surgical re-exploration but a higher incidence of AKI (19). Moreover, Frumento et al. found more strokes in patients treated with half-dose aprotinin than those treated with full-dose aprotinin (20). Therefore, it is currently difficult to recommend a regimen over the other.

Our study presented several limits that must be addressed. First, it is a retrospective cohort study, including two different periods of treatment, and it is therefore at high risk of bias in the selection of patients. Nevertheless, it seems that patients included in the aprotinin group presented higher risk factors for bleeding (lower platelet count and fibrinogen plasma level, trend in higher POBS-Card scores). The fact that aprotinin remains associated with better outcomes may be strengthened by these differences in patient characteristics. Second, we evaluated a limited number of patients which limits the confidence in safety results. Regarding the mechanism of missing data, we cannot formally exclude that some data were missing not at random, particularly in the most critical emergency cases where complete preoperative assessment was not always feasible. We therefore acknowledge that the presence of missing data represents an additional limitation of our study. Third, we did not evaluate costs, but a recent French study by Colson et al. showed a saving of €3,136 from intervention to ICU discharge with aprotinin compared with the use of tranexamic acid, despite their difference in cost: €500 for half-dose aprotinin and €4 for tranexamic acid in this study (21). This was mainly due to fewer transfusions and a shorter ICU stay for the aprotinin group.

Conclusions

This single-center retrospective study suggests that aprotinin offers a significant reduction in perioperative bleeding severity compared with tranexamic acid in non-elective cardiac procedures involving the ascending aorta. Our data show that bleeding risk increases with the emergency level and prolonged CPB duration, underscoring the importance of aggressive hemostatic strategies in emergent and complex aortic interventions.

Our retrospective study with a small sample size does not allow us to draw a formal conclusion, and further prospective randomized studies are needed to clarify the efficacy and safety of aprotinin compared to tranexamic acid.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-0931/rc

Data Sharing Statement: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-0931/dss

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-0931/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-0931/coif). E.B. reports receiving payments from LFB Pharma. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. According to French regulations for retrospective observational studies using anonymized routine-care data, formal approval from an Institutional Review Board or Ethics Committee and written informed consent were not required. The data collection adhered to the MR004 methodology from the national committee for informatics and freedom (CNIL).

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Landis RC, Asimakopoulos G, Poullis M, et al. The antithrombotic and antiinflammatory mechanisms of action of aprotinin. Ann Thorac Surg 2001;72:2169-75. [Crossref] [PubMed]
2. Fergusson DA, Hébert PC, Mazer CD, et al. A comparison of aprotinin and lysine analogues in high-risk cardiac surgery. N Engl J Med 2008;358:2319-31. [Crossref] [PubMed]
3. Later AF, Maas JJ, Engbers FH, et al. Tranexamic acid and aprotinin in low- and intermediate-risk cardiac surgery: a non-sponsored, double-blind, randomised, placebo-controlled trial. Eur J Cardiothorac Surg 2009;36:322-9. [Crossref] [PubMed]
4. Khair S, Perelman I, Yates J, et al. Exclusion criteria and adverse events in perioperative trials of tranexamic acid in cardiac surgery: a systematic review and meta-analysis. Can J Anaesth 2019;66:1240-50. [Crossref] [PubMed]
5. Ahn KT, Yamanaka K, Iwakura A, et al. Usefulness of intraoperative continuous infusion of tranexamic acid during emergency surgery for type A acute aortic dissection. Ann Thorac Cardiovasc Surg 2015;21:66-71. [Crossref] [PubMed]
6. Besnier E, Schmidely P, Dubois G, et al. POBS-Card, a new score of severe bleeding after cardiac surgery: Construction and external validation. JTCVS Open 2024;19:183-99. [Crossref] [PubMed]
7. De Hert S, Ouattara A, Royston D, et al. Use and safety of aprotinin in routine clinical practice: A European postauthorisation safety study conducted in patients undergoing cardiac surgery. Eur J Anaesthesiol 2022;39:685-94. [Crossref] [PubMed]
8. Nashef SA, Roques F, Sharples LD, et al. EuroSCORE II. Eur J Cardiothorac Surg 2012;41:734-45.
9. Alghamdi AA, Davis A, Brister S, et al. Development and validation of Transfusion Risk Understanding Scoring Tool (TRUST) to stratify cardiac surgery patients according to their blood transfusion needs. Transfusion 2006;46:1120-9. [Crossref] [PubMed]
10. Wahba A, Kunst G, De Somer F, et al. 2024 EACTS/EACTAIC/EBCP Guidelines on cardiopulmonary bypass in adult cardiac surgery. Br J Anaesth 2025;134:917-1008. [Crossref] [PubMed]
11. Ali JM, Gerrard C, Clayton J, et al. Hemostasis Checklist Reduces Bleeding and Blood Product Consumption After Cardiac Surgery. Ann Thorac Surg 2021;111:1570-7. [Crossref] [PubMed]
12. Dyke C, Aronson S, Dietrich W, et al. Universal definition of perioperative bleeding in adult cardiac surgery. J Thorac Cardiovasc Surg 2014;147:1458-1463.e1. [Crossref] [PubMed]
13. Casselman FPA, Lance MD, Ahmed A, et al. 2024 EACTS/EACTAIC Guidelines on patient blood management in adult cardiac surgery in collaboration with EBCP. Eur J Cardiothorac Surg 2025;67:ezae352. [Crossref] [PubMed]
14. Gallo E, Gaudard P, Provenchère S, et al. Use of Aprotinin versus Tranexamic Acid in Cardiac Surgery Patients with High-Risk for Excessive Bleeding (APACHE) trial: a multicentre retrospective comparative non-randomized historical study. Eur J Cardiothorac Surg 2024;65:ezae001. [Crossref] [PubMed]
15. Shi J, Zhou C, Pan W, et al. Effect of High- vs Low-Dose Tranexamic Acid Infusion on Need for Red Blood Cell Transfusion and Adverse Events in Patients Undergoing Cardiac Surgery: The OPTIMAL Randomized Clinical Trial. JAMA 2022;328:336-47. [Crossref] [PubMed]
16. Reidy B, Aston D, Sitaranjan D, et al. Lack of efficacy of aprotinin over tranexamic acid in type A aortic dissection repair. Transfusion 2024;64:846-53. [Crossref] [PubMed]
17. Dietrich W, Busley R, Boulesteix AL. Effects of aprotinin dosage on renal function: an analysis of 8,548 cardiac surgical patients treated with different dosages of aprotinin. Anesthesiology 2008;108:189-98. [Crossref] [PubMed]
18. Levy JH, Salmenpera M, Bailey JM, et al. Is there a difference between full dose and half dose aprotinin in chest tube drainage following cardiac surgery? Anesthesiology 1992;77:A100.
19. Colson PH, Gaudriot B, Provenchère S, et al. Half-Dose versus Full Dose of Aprotinin in Cardiac Surgery: A Post-hoc Analysis of the Aprotinin European Registry. Eur J Cardiothorac Surg 2025;67:ezaf260. [Crossref] [PubMed]
20. Frumento RJ, O'Malley CM, Bennett-Guerrero E. Stroke after cardiac surgery: a retrospective analysis of the effect of aprotinin dosing regimens. Ann Thorac Surg 2003;75:479-83; discussion 483-4. [Crossref] [PubMed]
21. Colson P, Fellahi JL, Gaudard P, et al. Cost Analysis of Aprotinin Reintroduction in French Cardiac Surgery Centres: A Real-World Data-Based Analysis. Adv Ther 2023;40:1803-17. [Crossref] [PubMed]