Pharmacologic Hemostasis by Preoperative Recovery of Coagulation

Potent preoperative anticoagulants and antiplatelet drugs frequently increase bleeding and transfusion significantly after CABG. Therefore, when clinically feasible, they should be discontinued preoperatively to allow recovery of the coagulation system (Class IIb recommendation; Level C evidence). The timing of discontinuation depends on the half-life of the particular agent and the possibility of reversibility. The exception to this principle is unfractionated heparin, which may be discontinued shortly before CABG or not at all.

High-intensity platelet blockade with thienopyridines such as clopidogrel may be associated with life-threatening bleeding after CABG. It is reasonable to discontinue this potent platelet blockade before elective surgery to limit blood loss and transfusion (Class I recommendation; Level B evidence). The period of discontinuation is dependent on the properties of the drug, but a period of at least 5 days is recommended for clopidogrel and ticagrelor. However, based on the variability in response and resistance to common antiplatelet agents, the period of discontinuation may be as short as 3 days. The use of point-of-care testing for platelet adenosine diphosphate (ADP) responsiveness may be reasonable for identifying nonresponders and those eligible for earlier CABG; however, strong evidence is lacking for this recommendation (Class IIb recommendation; Level C evidence). The recommendation for prasugrel is for the drug to be stopped at least 7 days before planned surgery. In the presence of coronary stents, whether bare-metal or drug-eluting stents, early withdrawal of antiplatelet therapy can precipitate stent thrombosis. The options to maintain stent patency must be considered, including preoperative hospitalization to substitute thienopyridine therapy with short-acting intravenous platelet blockade. Cangrelor is a promising new short-acting, intravenous P2Y12 inhibitor that has already shown maintenance of platelet inhibition when used as a bridging agent without a significant increase in operative bleeding when compared with placebo.

It is reasonable to stop low-intensity antiplatelet therapy (e.g., aspirin) preoperatively in elective patients without acute coronary syndromes to reduce blood loss and transfusion after CABG (Class IIa recommendation; Level A evidence). In the setting of emergent CABG, aspirin should be continued because the small bleeding risk is outweighed by its overall benefits (Class IIa recommendation; Level A evidence).

Limiting Bleeding and Transfusion with Autologous Donation and Erythropoietin

Preoperative autologous blood donation is reasonable in selected patients, especially when combined with appropriate erythropoietin and iron therapy (Class IIa recommendation; Level A evidence). Although common in noncardiac surgical cases, cardiac surgery does not routinely use this technique because of concerns about an increased incidence of myocardial infarction before surgery in CABG patients. However, autologous donation can be used safely before elective CABG, and this practice is associated with a significant reduction in allogeneic blood transfusion. The use of recombinant human erythropoietin to restore red cell volume in those patients undergoing autologous donation should be considered, but it should be balanced with the potential for thrombotic cardiovascular events in this population (Class IIb recommendation; Level A evidence). Preoperative erythropoietin, plus iron, given several days before surgery is also indicated for boosting red blood cell mass in anemic patients, in patients who refuse allogeneic blood transfusion (e.g., Jehovah’s Witness patients), or in patients who are at high risk of postoperative anemia (Class IIa recommendation; Level B evidence). Given the evidence that preoperative anemia independently predicts death, stroke, and renal failure after CABG, randomized trials with preoperative erythropoietin to augment red blood cell mass have already shown effectiveness in decreasing the incidence of postoperative transfusion and increasing postoperative hemoglobin values.

Pharmacologic Hemostasis with Antifibrinolytic Agents

Activation of the fibrinolytic cascade in cardiac surgery patients contributes to bleeding. Therefore antifibrinolytic agents were introduced as a pharmacologic technique aimed at improving hemostasis. Antifibrinolytics are the most extensively studied blood conservation agents, and they have been shown to decrease postoperative bleeding and the need for allogeneic blood transfusion in CABG patients, as well as other cardiac and noncardiac surgical cases. The antifibrinolytic agent aprotinin was withdrawn from the market in late 2007 because of concerns about patient safety when used in cardiac surgery. The Blood Conservation Using Antifibrinolytics in a Randomized Trial (BART) study was suspended for the apparent increase in mortality rate caused by aprotinin as compared with tranexamic acid and aminocaproic acid. Even before this development, safety concerns related to anaphylaxis and renal dysfunction had already significantly limited the clinical application of aprotinin. Two further massive outcome analyses of aprotinin in CABG (cumulative n = 88,474) have also documented a significant increase in mortality rate in CABG patients exposed to perioperative aprotinin as compared with aminocaproic acid. Given this information and despite the fact that aprotinin decreases postoperative bleeding and transfusion, the use of aprotinin for CABG receives a Class III recommendation (Level A evidence) because risks outweigh benefits. Since its removal from the market, further analyses and newer data suggest that aprotinin may continue to have a role in cardiac surgery, particularly in patients who are at high-risk of bleeding.

The remaining available antifibrinolytics are the lysine analogs, tranexamic acid and aminocaproic acid. High-quality meta-analyses consistently support the safety and efficacy of the lysine analogs for blood conservation in CABG surgery. These agents significantly reduce bleeding and blood component transfusion across multiple randomized trials. There are reports of increased occurrence of seizures with tranexamic acid use, but this has not been confirmed by large randomized trials. Based on the available data, the application of these agents, particularly in high-risk CABG subgroups, has received a Class I recommendation (Level A evidence).

Pharmacologic Hemostasis with Desmopressin and Recombinant Factor VIIa

Desmopressin acetate is a synthetic analog that releases factor VIII precursors and von Willebrand factor from vascular endothelium. Desmopressin therapy is not unreasonable to attenuate excessive bleeding in patients with platelet dysfunction secondary to uremia, cardiopulmonary bypass, and type I von Willebrand disease (Class IIb recommendation; Level B evidence). Furthermore, preoperative platelet dysfunction detectable by point-of-care testing can often be reversed by desmopressin therapy. Thus desmopressin is indicated perioperatively in selected cases with evidence of platelet dysfunction. However, routine prophylactic desmopressin does not reduce bleeding and transfusion after CABG (Class III recommendation; Level A evidence).

Recombinant factor VIIa (rFVIIa) therapy has demonstrated efficacy in the management of nonsurgical bleeding after CABG. This efficacy is based on a consistent trend from multiple case series that have been systematically reviewed. Although rFVIIa has been shown to decrease reoperation and blood transfusion in a randomized trial, concerns regarding adverse events such as thrombotic complications limit routine use. Until larger randomized controlled trials become available to further evaluate the safety and efficacy of this intervention, rFVIIa should be considered only in those CABG patients with massive and refractory nonsurgical bleeding that has not responded to routine hemostatic treatment options (Class IIb recommendation; Level B evidence).

Limiting Bleeding and Transfusion with Avoidance of Cardiopulmonary Bypass

Because cardiopulmonary bypass (CPB) is associated with hemostatic disturbances, it is not surprising that CABG without CPB was associated with decreased bleeding and transfusion when compared with CABG plus CPB in a meta-analysis of randomized trials. Off-pump CABG is reasonable for blood conservation, provided that emergent conversion to on-pump bypass is unlikely based on surgeon experience or patient characteristics (Class IIa recommendation; Level A evidence). Emergent conversion to CABG with CPB is associated with significantly greater bleeding and reoperation.

Limiting Bleeding and Transfusion with Modified Cardiopulmonary Bypass

The conduct of CPB may significantly affect bleeding and transfusion after CABG. The design of the CPB circuit is the first major consideration. The hemostatic possibilities in CPB hardware design include oxygenator design (bubble or membrane), pump type (centrifugal or roller), and circuit type (biocompatible and/or minimized low-prime). The second major consideration is anticoagulation management for CPB with heparin and protamine. The evidence and recommendations for each of these considerations are as follows.

A membrane oxygenator during CPB is not unreasonable for reducing blood utilization (Class IIb recommendation; Level C evidence). Membrane oxygenators have largely replaced bubble oxygenators in contemporary clinical practice because they are associated with reductions in cerebral emboli and blood transfusion. CPB pump design, however, has less of a role in perioperative blood conservation after CABG. All pump designs, whether centrifugal or roller, provide acceptable hemostatic performance. Despite theoretical advantages of the centrifugal design over the roller design such as less hemolysis and reduced complement activation, consistent clinical reductions in bleeding and transfusion after CABG have not been observed in randomized trials. However, it is not unreasonable to prefer centrifugal pumps for their enhanced safety elements (Class IIb recommendation; Level B evidence).

Biocompatible CPB circuits are not unreasonable for promoting blood conservation (Class IIb recommendation; Level A evidence). Because they attempt to mimic the endothelial surface by coating the bypass circuit with various compounds (such as heparin), these circuits have shown benefit in decreasing the inflammatory response and hemostatic activation associated with CPB. A large meta-analysis of more than 4000 patients concluded that biocompatible circuits are associated with a lower incidence of blood transfusion; however, they are best used in conjunction with other blood conservation techniques for the greatest benefit to be observed. Minicircuits have been shown to reduce hemodilution and should also be used for blood conservation, especially in patients with preoperative anemia (Class I recommendation; Level A evidence). These minimized circuits contain a reduced priming volume that limits hemodilution on initiation of CPB. Benefits are especially noted in pediatric patients or Jehovah’s Witness patients, but they can also be used with efficacy in CABG patients. Clinical trials have documented significant reductions in bleeding and transfusion after CABG with the low-prime CPB circuit as compared with conventional CPB. Furthermore, there is high-quality evidence that these beneficial outcomes are similar in magnitude to the hemostatic benefit from CABG without CPB.

Anticoagulation for CABG with CPB is used to limit cellular and coagulation factor activation and to prevent circuit thrombosis. Unfractionated heparin is the anticoagulant of choice because it is effective, reversible with protamine, generally well-tolerated, and inexpensive. The activated clotting time (ACT) is a standard point-of-care test to monitor heparin effect during CPB. An ACT time of greater than 400 seconds is the traditional standard for safe CPB, originally based on a 1978 primate study with bubble oxygenators. For reduction of the total protamine dosing needed for heparin reversal, the idea of low-dose heparin therapy, in conjunction with heparin-coated CPB circuits, has been evaluated. Aiming for an ACT of 300 seconds, the goals of this concept were to limit bleeding and transfusion after CABG by decreasing heparin exposure and the need for protamine reversal. This intervention was considered as not unreasonable for promoting blood conservation, but safety concerns such as thrombosis have not been well studied (Class IIb recommendation; Level B evidence). In contrast to lower heparin dosing, the use of high-dose heparin therapy during prolonged CPB (longer than 2 to 3 hours) has also been suggested for blood conservation. High-dose heparin therapy can decrease thrombin generation, fibrinolytic activity, and platelet activation. In a randomized study with point-of-care testing (heparin concentration and ACT) to maintain appropriate heparin concentrations for an ACT greater than 480 seconds, a reduction in blood product utilization was observed, likely secondary to preservation of the coagulation system. Considering this evidence, it is not unreasonable to use high-dose heparin therapy monitored with point-of-care testing to reduce hemostatic activation, platelet consumption, and need for transfusion in prolonged CPB cases (Class IIb recommendation; Level B evidence).

Heparin reversal with protamine can affect bleeding and transfusion after CABG with CPB because excess protamine is itself an anticoagulant. Protamine titration or empiric low-dose regimens not only lower the total protamine dose but also have been shown in clinical trials to reduce bleeding and transfusion but not consistently. One randomized trial found 80% fewer transfusions when patient response tests were incorporated into the management. However, another randomized trial noted no difference in postbypass hemostasis when titration methods were used. Results from other similar trials are also inconclusive regarding the benefits of this method. Although protamine titration or empiric low-dose protamine therapy is not unreasonable (Class IIb recommendation; Level B evidence), more consistent evidence of benefit is required before a higher class recommendation can be assigned.

Limiting Bleeding and Transfusion with Modified Blood Management

Conservation of the patient’s red cell volume with a multimodal approach is the first principle of modified blood management for limitation of bleeding and transfusion after CABG. Routine red-cell saving with centrifugation limits blood transfusion in CABG with CPB (Class I recommendation; Level A evidence). Because of safety concerns, this is not indicated in patients with infection (the concern is septicemia). A change from prior guidelines is that cell salvage from the operative field in patients with known malignancy should now be considered in high-risk patients (Class IIb recommendation; Level B evidence). This change comes from evidence that suggests worsened outcomes and increased recurrence rates in patients with malignancy when allogeneic blood is transfused. Intraoperative autotransfusion directly from cardiotomy suction or recycled from a cell-saving device is also not unreasonable to augment blood conservation (Class IIb recommendation; Level C evidence). Extensive cell-saving, however, leads to loss of coagulation factors and platelets, which may result in a bleeding diathesis. This deleterious effect of extensive cell-saving can be offset after CPB by some form of pump salvage and reinfusion of residual pump blood, which is considered a reasonable means of blood conservation (Class IIa recommendation; Level C evidence). Centrifugation, rather than direct infusion, of this residual pump blood is preferred (Class IIa recommendation; Level A evidence) because increased fibrin degradation occurs with direct infusion. This can lead to increased bleeding and transfusion requirements.

Acute normovolemic hemodilution is not unreasonable for blood conservation in cardiac surgery (Class IIb recommendation; Level B evidence). This often involves the removal of one to two units of autologous blood immediately before initiation of CPB. For the circulating blood volume to be maintained, the volume of removed blood is replaced 1 : 1 with crystalloid or colloid. An advantage of this technique, along with decreasing allogeneic blood use, is that platelet function and clotting factors are preserved in the autologous blood because it does not undergo the deleterious effects of CPB. This is especially beneficial for hemostasis when the blood is returned to the patient after CPB. Another advantage is that the blood can be used as a first source of volume should transfusion be needed while the patient is undergoing bypass. Although it is considered safe in several patient groups, patient groups with contraindications to this technique include unstable patients and those with preoperative anemia.

Retrograde autologous priming is an intervention for blood conservation that, similar to acute normovolemic hemodilution, is instituted just before initiation of CPB. The crystalloid prime volume of the arterial limb of the CPB circuit is cleared retrograde by back bleeding from the aortic cannula. Similarly, the venous limb is cleared in an antegrade manner. While recent trials have shown that this technique decreases hemodilution and the need for transfusions, others have demonstrated no significant benefit. Despite this limitation, retrograde autologous priming is not unreasonable for blood conservation after CABG (Class IIb recommendation; Level B evidence).

The use of platelet plasmapheresis is a reasonable strategy for conserving blood as part of a multimodality approach in high-risk patients (Class IIa recommendation; Level A evidence). Selective removal during continuous centrifugation results in platelet-rich-plasma, which can be returned to the patient after CPB for hemostasis because these platelets have been spared from CPB dysfunction. Although current evidence ranges from either no benefit to significant benefit in reducing bleeding and transfusion, platelet plasmapheresis should only be performed if an adequate platelet yield can be reliably obtained.

The presence of leukocytes in packed red cells can lead to inflammatory effects and the potential for infection. Therefore, if allogeneic blood is needed, it is reasonable to use leukoreduced donor blood if available (Class IIa recommendation; Level B evidence). Leukocyte filtration during CPB may theoretically improve bleeding and transfusion after CABG. However, clinical trials of this intervention have failed to show consistent hemostatic benefit after CABG. Furthermore, there is evidence that leukocyte depletion during CPB may activate white cells. For these reasons, the use of leukocyte filters during CPB are not indicated for blood conservation in CABG (Class III recommendation; Level B evidence).

Modified ultrafiltration (MUF) is a form of ultrafiltration that removes water and inflammatory mediators from blood after completion of CPB. This results in less hemodilution. A large meta-analysis of 1004 patients revealed that MUF decreases postoperative bleeding and transfusion requirements in patients undergoing cardiac surgery. Consequently, MUF is now indicated for blood conservation in patients undergoing CABG with CPB (Class I recommendation; Level A evidence).

Although postoperative transfusion of shed mediastinal blood may limit blood transfusion, multiple clinical trials have failed to demonstrate consistent benefit. Furthermore, there is potential for harm, including sternal and systemic infection. Given the lack of consistent clinical benefit and evidence of harm, direct infusion of shed mediastinal blood from postoperative chest tube drainage is not indicated for perioperative blood conservation after CABG (Class III recommendation; Level B evidence).

Another principle of modified blood management for limitation of bleeding and transfusion after CABG is a perioperative transfusion protocol to standardize institutional transfusion practice as far as possible. Based on expert opinion and consensus, the following recommendations all relate to this principle, and they can guide the decision of whether to transfuse.

It is reasonable to transfuse hemostatic blood products based on clinical evidence of bleeding, preferably guided by point-of-care testing (Class IIa recommendation; Level C evidence). In CPB, it is not unreasonable to maintain the hemoglobin level at 7 g/dL or greater in patients with a risk of critical end-organ injury (Class IIb recommendation; Level C evidence). Transfusion is not recommended for a hemoglobin concentration greater than 10 g/dL (Class III recommendation; Level C evidence), except in patients with critical noncardiac end-organ ischemia, in which case it is not unreasonable to maintain the hemoglobin concentration at greater than 10 g/dL (Class IIb recommendation; Level C evidence).

Mechanical Hemostasis with Positive End-Expiratory Pressure

Positive end-expiratory pressure (PEEP) exerts mechanical pressure on the heart and so may limit bleeding after CABG. Two clinical studies with no control group have documented control of excessive bleeding with escalating levels of PEEP up to a maximum of 20 cm H ₂ O. A trial of therapeutic PEEP to ameliorate excessive bleeding is not unreasonable (Class IIb recommendation; Level B evidence). In those cases in which the use of PEEP is effective, a reduction in postoperative bleeding often becomes apparent within an hour of initiation. When significant mediastinal bleeding is not already apparent, the use of prophylactic PEEP does not reduce postoperative bleeding (Class III recommendation; Level B evidence). It is also important to keep in mind the risks of cardiovascular compromise after CABG with escalating levels of PEEP.