Antifibrinolytic Drugs

Antifibrinolytic drugs have been studied extensively for use in cardiac surgery as well as in several other surgical populations that are at risk of bleeding. Epsilon aminocaproic acid (EACA) and tranexamic acid (TXA) are synthetic lysine analogs that bind competitively to plasmin and plasminogen, preventing their binding to and breakdown of fibrin ( Figure 24-1 ). Both of these drugs undergo renal excretion and concentration, requiring dose adjustment for patients with renal insufficiency. TXA is 10 times more potent than EACA in terms of affinity for the lysine binding site.

Mechanisms of Antifibrinolytic Drugs. A, Fibrinolysis inhibited by binding of tranexamic acid (TA) or epsilon aminocaproic acid (EACA). B, Aprotinin (AP) inhibits fibrinolysis by inhibiting both kallikrein and plasmin. CPB, cardiopulmonary bypass.

Aprotinin is a nonspecific serine protease inhibitor derived from bovine lung. It acts at several proteases, including plasmin, kallikrein, trypsin, and factor XII (see Figure 24-1 ). Compared with the lysine analogs, aprotinin not only inhibits fibrinolysis but also complement activation and contact activation of both coagulation and inflammation. In addition, aprotinin also preserves platelet function after cardiopulmonary bypass. There is a small risk of anaphylaxis, especially with repeated exposure. It should be noted that aprotinin artificially prolongs celite-based activated clotting time (ACT) measurements; therefore a kaolin ACT test should be used. Dosing is based on kallikrein-inhibiting units (KIU). Aprotinin was withdrawn from the U.S. market in 2008 after Blood Conservation Using Antifibrinolytics in a Randomized Trial (BART) showed an increase in the mortality rate compared with the lysine analogs (see further on).

Desmopressin

Desmopressin (1-deamino-8-D-arginine vasopressin, DDAVP) is a synthetic analog of the hormone arginine vasopressin, also known as antidiuretic hormone (ADH). It is Food and Drug Administration (FDA)-approved for the treatment of hemophilia A (when factor VIII activity is greater than 5%), von Willebrand disease type 1, and diabetes insipidus. It has also been used off-label to treat other forms of platelet dysfunction (e.g., uremia-induced or postcardiopulmonary bypass) because of its ability to release endogenous stores of factor VIII, von Willebrand factor (vWF), and plasminogen activator, which in turn enhance platelet function. Effects are seen within 30 minutes of intravenous (IV) administration. Contraindications include moderate to severe renal insufficiency and hyponatremia. In the perioperative setting, transient hypotension due to decreased systemic vascular resistance is the most common side effect, although this is mitigated by slow infusion.

Protamine

Protamine is a strongly basic polypeptide used in the reversal of unfractionated heparin. It binds to the highly acidic heparin molecules to form a stable salt that lacks anticoagulant properties. However, protamine by itself is a weak anticoagulant, with effects on factor V, platelets, and fibrinolysis. Inadequate or excess doses can both lead to excess bleeding. It is not effective for reversal of low-molecular-weight heparin. Reactions may include histamine release and anaphylactic, anaphylactoid, and pulmonary vasoconstriction. Slow administration can help prevent some of these reactions. Treatment of a protamine reaction is supportive.

Vitamin K

Warfarin works by inhibiting the vitamin K-dependent gamma-carboxylation of factors II, VII, IX, and X, as well as proteins C and S. As such, warfarin can be reversed by the administration of vitamin K, and effects are seen in 4 to 6 hours if given intravenously. Oral administration requires up to 24 hours for full effect. If more urgent reversal is necessary, fresh-frozen plasma (FFP) or prothrombin complex concentrates (PCCs) should be used (see next section), but vitamin K should still be administered because of the short half-life of exogenous factors. The American College of Chest Physicians recommends the addition of 5 to 10 mg IV vitamin K be given in addition to plasma transfusion for rapid reversal. Concerns over anaphylaxis have led many to avoid IV use of vitamin K, but these reactions are quite rare, and IV administration should not be avoided if urgent reversal is needed.

Prothrombin Complex Concentrates

PCCs are isolated from pooled human plasma and contain varying amounts of the vitamin K-dependent clotting factors. They may also contain varying amounts of proteins C and S, antithrombin, and heparin. PCCs can be divided into two groups based on whether they contain significant amounts of factor VII. Those that do not are considered 3-factor PCCs, whereas those that do are 4-factor. In the United States, 3-factor PCCs are approved for the treatment of bleeding in patients with hemophilia B, and 4-factor PCCs are currently not approved at all. Recombinant factor IX products are available and are the mainstay of treatment for hemophilia B, leaving PCCs to be used off-label as an alternative to FFP. Compared with FFP, the theoretical advantages of PCCs include immediate availability without thawing, small volume of administration, rapid administration, and viral inactivation. Disadvantages include the potential for thromboembolic events, exposure to multiple donors, and cost. In addition, FFP contains more than just the specific factors found in PCCs (e.g., fibrinogen). PCCs are labeled based on their factor IX content, which can lead to confusion with purified factor IX products that do not contain any other factors.

Recombinant Activated Factor VII

Recombinant activated factor VII (rFVIIa) is FDA approved for the treatment of bleeding in congenital factor VII deficiency and in hemophilia patients with inhibitors to factor VIII or IX. It has been used off-label in a variety of scenarios, including trauma, intracranial hemorrhage, surgery, and reversal of anticoagulation. The mechanism of action is related to its ability to complex with tissue factor, allowing it to activate factors X and IX, which in turn complex with other factors to convert prothrombin to thrombin. It should be noted that rFVIIa is a relatively expensive drug, costing approximately $10,000 for a typical adult dose ( Table 24-2 ).

Antifibrinolytic Drugs

The landmark study BART was the first large-scale trial (n = 2331) to compare aprotinin with the lysine analogs in a head-to-head fashion. It focused on patients at high-risk of bleeding, which is the population in which aprotinin was thought to be most beneficial. The trial was stopped early because of a strong trend toward a higher mortality rate in the aprotinin group—an absolute risk increase of 2.1% and a relative risk increase of 54% compared with the lysine analogs. The investigators noted a statistically significant increase in postoperative creatinine levels but only a trend toward an increased need for renal replacement therapy that did not reach statistical significance. On the basis of these data, aprotinin was withdrawn from the market in 2008. BART confirmed what had been reported in other observational studies.

The Cochrane Collaboration recently updated their meta-analysis of antifibrinolytic drugs. They evaluated 252 trials, of which 173 involved cardiac surgery. Compared with placebo, aprotinin reduced the relative rate of blood transfusion by 32%, reduced the intraoperative blood loss by 148 mL, reduced the postoperative blood loss by 370 mL, and reduced the relative risk of reoperation for bleeding by 54%. There was no statistically significant increase in mortality rates, myocardial infarction, stroke, deep venous thrombosis, or pulmonary embolus. There was a trend toward an increased risk of renal dysfunction, but this was not statistically significant.

This meta-analysis also evaluated aprotinin compared with TXA and EACA. Aprotinin was more effective at reducing postoperative blood loss, the rate of transfusion, and the need for reoperation for bleeding. There was again no statistically significant difference in myocardial infarction, stroke, and renal failure. The investigators did, however, note an increase in the mortality rate, similar to BART. Of course, the meta-analysis itself was heavily influenced by BART itself because it was the largest in this field.

Although TXA is 10 times more potent than EACA, it was not more effective when compared head-to-head. The previously mentioned meta-analysis also came to the conclusion that there was no significant difference in efficacy between the lysine analogs, although investigators did note that there were significantly more data available for TXA. EACA is considerably less costly than TXA in the United States, although this may not be the case in other countries. Accordingly, a recent survey showed that EACA is used at most institutions in the United States, whereas TXA is the drug of choice in Canada.

The coagulopathy of traumatic injury is complex and multifactorial, but hyperfibrinolysis appears to play a key role. The highly publicized Clinical Randomization of an Antifibrinolytic in Significant Haemorrhage-2 (CRASH-2) trial compared the use of TXA with placebo in 20,211 trauma patients with (or at risk of) significant bleeding. The all-cause mortality rate was reduced by 9% (from 16% to 14.5%), and deaths due to bleeding were reduced by 15% (from 5.7% to 4.9%) without any increase in thrombotic events. Interestingly, no significant reduction in transfusion rate was found, which leaves the protective mechanism of TXA unclear. Further analysis showed that the results after early administration (within 3 hours) were even more impressive, but late administration was actually harmful. Although the relative risk reduction is more impressive than the absolute risk reduction, it is important to keep in mind that TXA was one of the few interventions found to be useful in a recent systematic review of the management of hemorrhage in trauma.

Liver transplantation has been associated with a high risk of blood loss and transfusion, and worse outcomes are reported in patients receiving allogenic blood transfusions. The effects of liver failure on the coagulation system are complex in that a delicate balance exists between hypercoaguability and hypocoaguability that may vary from patient to patient. Hyperfibrinolysis during reperfusion of the new graft has long been recognized as a problem, prompting several trials of antifibrinolytic therapy. A 2011 meta-analysis included eight trials with aprotinin, five with TXA, and one with EACA. In general, the trend was toward a reduction in blood loss that did not reach statistical significance. However, only aprotinin demonstrated a statistically significant reduction in the number of products transfused to each patient. Although there are concerns that antifibrinolytics may tip the balance toward a hypercoaguable state, no differences were seen in mortality rates, graft failure, or thromboembolic events. A retrospective study specifically found a trend toward increased arterial and venous thrombosis when aprotinin was used, but it did not reach statistical significance.

Limited data suggest that antifibrinolytics are useful for reducing blood loss and transfusion requirements in hepatic resection, and morbidity and mortality rates are not increased.

Among orthopedic procedures, total joint replacement and spine surgery are associated with the most blood loss. Two recent meta-analyses looked at the use of TXA in total hip and knee replacement surgery, respectively. Both showed that TXA was effective in reducing blood loss and transfusion requirements, without an increased risk of thromboembolic events. In total hip replacement, total blood loss was reduced by an average of 289 mL, and the transfusion rate was reduced by 20%. The effects were even more profound in total knee replacement: the reduction in total blood loss was 591 mL, and the transfusion rate was reduced by 39%. These effects seemed to be dose-dependent, but there are insufficient data to draw firm conclusions. The results are similar for aprotinin and EACA, but the data for EACA are limited; only three trials were included in a general meta-analysis of antifibrinolytic drugs. Use in spine surgery has also been successful, although the data are not as robust as in total joint replacement.

Postpartum hemorrhage (PPH) is a leading cause of maternal death in obstetrics. Management has generally focused on uterotonic drugs, such as oxytocin, prostaglandins, and ergometrine. TXA has been used when these drugs are insufficient since at least 1996. A meta-analysis of three randomized trials showed that prophylactic use of TXA reduced the incidence of PPH by 56%. More recently, a randomized trial evaluated the use of TXA after PPH was diagnosed. The median blood loss was reduced by 173 mL, the duration and severity of bleeding were reduced, and fewer transfusions were required. The reduction in transfusions included hemostatic products, such as FFP, platelets, and cryoprecipitate. This study was not powered to evaluate safety, but no major adverse events occurred.

The use of antifibrinolytics in pediatric surgery remains somewhat controversial because of limited data and variable dosing of the drugs. A 2009 meta-analysis included 23 trials in cardiac surgery and five trials in scoliosis surgery. In cardiac surgery, TXA reduced blood loss by an average of 11 mL/kg compared with placebo, but no conclusion could be made for EACA and aprotinin because of the heterogeneous nature of the data. Aprotinin reduced the volume of red blood cell transfusion by 4 mL/kg compared with 7 mL/kg for TXA. Similar reductions were seen for FFP. In scoliosis surgery, blood loss was reduced by an average of 385 mL by aprotinin and 682 mL by TXA. TXA also reduced the volume of red blood cell transfusion by 349 mL. A systemic review from 2008 that focused on pediatric heart surgery noted that patient populations and dosing regimens were highly variable between studies, which made comparisons difficult. The authors concluded that the benefit was likely highest in high-risk patients, such as those with cyanosis and those undergoing complex or revision surgery. Recently, TXA has also been shown to be effective in craniosynostosis surgery.

Desmopressin

Interest in the use of DDAVP in cardiac surgery increased significantly after one early trial demonstrated a reduction in blood loss by 40% in patients without platelet defects known to respond to this drug. Although the study was a randomized, double-blinded one, it included only 70 patients. Unfortunately, subsequent studies have been unable to substantiate such a large effect.

A 2008 meta-analysis on desmopressin included 42 trials, 28 of which were in cardiac surgery. Most of these trials were relatively small and had limited follow-up. Overall, there was a decrease in blood loss by 80 mL per patient and a decrease in blood transfusion by 0.3 units per patient. A trend toward a reduction in platelet transfusion was not statistically significant. Results were similar in both cardiac and noncardiac surgical populations. No significant difference was found in rates of mortality, myocardial infarction, stroke, or reoperation for bleeding. Transient hypotension was the most common adverse event. An earlier meta-analysis focusing on cardiac surgical patients found a small decrease in blood loss and units of blood transfused, especially when cardiopulmonary bypass times exceeded 140 minutes, but no statistically significant benefit was seen in patients taking aspirin.

Three studies have evaluated the targeted use of DDAVP in cardiac surgery based on platelet dysfunction identified by point-of-care testing. The first showed a significant reduction in blood loss and blood transfusions. The second study demonstrated that DDAVP is capable of increasing platelet function, but the authors did not evaluate the clinical impact. It is known that patients with severe aortic stenosis can develop type 2A von Willebrand disease. When decreased platelet function was detected by a point-of-care platelet function analyzer, DDAVP was able to reduce the platelet defect and reduce blood loss during aortic valve replacement.

DDAVP improved laboratory test results in patients undergoing hepatectomy in one study, but was unable to reduce blood loss or transfusion rates. DDAVP showed promise in one early study of spinal fusion, but this has not been confirmed in later studies.

Protamine

Several studies have shown that excess protamine after cardiopulmonary bypass can be detrimental to clotting function. Early studies showed that a ratio in excess of 1.3 mg protamine to 100 units of circulating heparin (the lowest tested) prolonged ACT and impaired platelet function. A more recent study using more sensitive tests demonstrated that ratios in excess of 1 mg protamine to 100 units circulating heparin were detrimental. Precise protamine dosing based on the measured circulating heparin level has been shown to reduce blood product use in cardiac surgery. Heparin rebound can occur 1 to 6 hours after neutralization but may be prevented with a low-dose protamine infusion (25 mg/hr for 6 hours) or treated with small additional doses of protamine (5 to 15 mg). Despite this evidence, many continue to administer too much protamine based on the total heparin dose.

Prothrombin Complex Concentrates

The majority of the evidence related to PCCs comes from retrospective studies and case reports, most of which were focused on the rapid reversal of warfarin anticoagulation. In addition, most of these studies are from Europe, where 4-factor PCCs are readily available. Compared with FFP, 4-factor PCCs are able to more rapidly and more completely adjust the international normalized ratio (INR) in patients with intracranial hemorrhaging taking warfarin. Similar results were seen in a study of cardiac surgical patients.

One study demonstrated that a 3-factor PCC alone was not as effective as FFP alone in normalizing the INR, but a higher success rate was achieved when a small amount of FFP was combined with the PCC. This was thought to be due to the lack of factor VII in the PCC, which could be provided by even a small amount of FFP. A similar study showed more rapid INR reversal with a combination of FFP and 3-factor PCC and a decreased incidence of volume overload compared with FFP alone. Two recent studies showed that patients with a higher initial INR were less likely to respond to a 3-factor PCC. Patients with a lower INR likely have higher circulating levels of factor VII and are therefore more capable of a response to a PCC lacking factor VII.

A meta-analysis sought to evaluate the safety of both 3- and 4-factor PCCs in the setting of warfarin reversal, finding that there was a low incidence of thromboembolic events. The trend was toward more events with 4-factor PCCs, but this was not statistically significant. The authors noted that their analysis was limited by the lack of randomized controlled trials in this area.

For these reasons and the previously discussed theoretical benefits of PCC over FFP, the American College of Chest Physicians recommends the use of 4-factor PCCs for life-threatening bleeding in patients taking warfarin.

Recombinant Activated Factor VII

rFVIIa has been used for a variety of off-label indications, and numerous case reports and small studies have supported these uses. A 2010 meta-analysis focusing on the safety of rFVIIa included 4468 patients in 35 placebo controlled trials. A significant increase in arterial thromboembolic events was seen, including a 2.6 times higher rate of coronary artery events. An earlier meta-analysis that focused primarily on patients with hemophilia found a low incidence (1% to 2%) of thromboembolic events, suggesting that the drug is safe when used for its approved indication.

Yet another meta-analysis focusing on off-label uses analyzed the results by patient population. In cardiac surgery, there were two randomized controlled trials (RCTs) and four observational studies. There was no survival benefit but an increase in thromboembolic events. A similar effect was seen in four RCTs and one observational study of intracerebral hemorrhage. Four RCTs and three observational studies in trauma patients demonstrated reductions in transfusion requirements and acute respiratory distress syndrome but no survival benefit. Interestingly, the thromboembolic risk was not increased. In liver transplantation, four RCTs and one observational study showed no effect on survival or thromboembolism. There was, however, a trend toward a reduction in packed red blood cell transfusion.