Chapter 13. Transfusion Therapy and Blood Conservation

With the development of cardiac surgery in the 1950s to correct congenital heart defects came the need for large-volume blood transfusions. In the 1960s and 1970s, the introduction of valve prostheses and direct grafting of coronary arteries made the correction of acquired heart disease a possibility. These landmarks, along with the liberal use of homologous blood transfusion therapy, led to rapid growth of the field. Commensurate with the growth of cardiac surgery as a field was an increasing incidence of transfusion-transmitted hepatitis in the 1970s, ultimately alerting the public and treating physicians to the concept of blood conservation. The emergence of infection by human immunodeficiency virus (HIV) greater heightened the interest in this area, leading to the current practices of blood conservation therapy in cardiac surgery.

Historically, open-heart surgery has been associated with a high usage of blood transfusion. Some reports suggest that up to 70% of this patient population requires blood transfusions, resulting in an average of two to four donor exposures per patient.1,2 It has been reported that 10% of all red blood cell units transfused in the United States are administered during coronary bypass surgery.3 Almost all patients received blood transfusion in the early days of cardiac surgery. However, with an increased awareness of blood-borne infectious diseases, lack of donors, great cost to both the patient and the institution, allergic reaction, blood-type mismatch, and the needs of special populations such as Jehovah's Witnesses, a greater effort has been made to perform open-heart procedures without blood transfusions even in high-risk patients. Advances in perioperative medications that minimize blood loss; greater tolerance of lower hematocrits, especially on bypass; and improvements in surgical techniques resulting in shorter operative times have allowed for these extensive procedures to be performed without significant blood loss.

The high transfusion rates associated with cardiac surgery have been well characterized and are likely caused by the coagulopathy, platelet dysfunction, and red cell hemolysis that occur as a result of the cardiopulmonary bypass circuit.4–6 The introduction of hemodilution using crystalloid pump-priming solution rather than whole blood dramatically reduced the transfusion requirements seen during coronary artery bypass grafting (CABG) procedures.7 Although this technique has reduced the amount of blood transfused during cardiopulmonary bypass (CPB), the resulting hemodilution contributes to the risk of low intraoperative and postoperative hematocrit, especially in patients who weigh less than 70 kg, thereby posing a new risk for transfusion.

Efforts at reducing the use of homologous blood in cardiac surgery began almost 40 years ago. The efforts to decrease allogeneic blood exposures have been a topic of constant review and attention because of the desires of both patients and their physicians to conserve blood during the perioperative period. These joint efforts have affected virtually every aspect of the manner in which heart surgery and CPB are performed. Our experience combined with the experiences of others has led to the development of an integrated, comprehensive blood conservation program that makes the goal of bloodless heart surgery possible.

Since the earlier edition of this chapter, aprotinin has been discontinued; however, we still include information on it for historical perspective. Also transfusion-related acute lung injury (TRALI) has become highly investigated. Review of our practice indicates that the timing of transfusion, or when packed red blood cells are given to treat anemia is of importance in cardiac surgery outcomes. We have also noticed that in our patient population, an academic medical center setting, baseline hematocrit has trended lower over the past several years. Conservation of blood products becomes even more critical as both transfusion risk and product demand increases.

Prediction of Bleeding

Avoidance of transfusion is desired for cardiac surgery patients. There are, however, many factors that cause patients to be either anemic or have excessive postsurgical bleeding requiring homologous blood transfusion. These mechanisms for bleeding include inherited and acquired disorders, including platelet dysfunction, coagulation factor deficiency, excessive fibrinolysis, and other induced alterations related to hypothermia or medications.8

Even with anemic preoperative patients we have been able to successfully avoid transfusion. This can be accomplished by reducing hemodilution via use of low prime perfusion circuitry and avoidance of excessive crystalloid infusion perioperatively. Current prime volume can safely be reduced to 700 mL. Retrograde autologous priming of the entire circuit further reduces perfusion-related hemodilution to negligible levels.

There have been several reports of minimizing the use of blood and blood products both intraoperatively and postoperatively. Although most of these studies have focused on the use of one particular modality or pharmacologic agent, these techniques were applied in the context of an entire set of blood conservation measures. Collectively the results of these reports provide an important body of information that can be used as an aid to evaluate the relative effectiveness of various combinations of the presently available blood conservation techniques.9–11

Later groups took advantage of pharmacologic developments such as serine protease inhibitors, antifibrinolytics, and erythropoietin. By the early 1990s, these pharmacologic adjuncts to blood conservation had become readily available and could be categorized as agents useful in perioperative stimulation of bone marrow for red blood cell production (eg, erythropoietin) and agents useful in reducing postoperative bleeding (eg, serine protease inhibitors and antifibrinolytics).

In 1990, Ovrum and colleagues confirmed the effectiveness of the simple “core” approach to blood conservation.11 In 121 consecutive elective CABG patients, the authors achieved a transfusion rate of 4.1% and 0.06 unit per patient. In 1991, Ovrum applied these same principles to 500 elective CABG patients and obtained similar low rates of transfusion (2.4% of patients).12 The authors found this six-step blood conservation program to be simple, safe, and cost effective.

By the 1990s, there also were some technical advances in CPB designed to minimize the need for transfusions as well as for hemodilution. With better oxygenators came a reduction in the volume of the CPB circuit, accompanied by a reduction in the amount of hemodilution that occurred.13 Circuit volume could be further reduced by replacing as much of the crystalloid circuit prime as possible with autologous blood drained from the arterial cannula into the circuit immediately prior to CPB, called retrograde autologous priming (RAP), as well as displacing the venous side of the crystalloid prime when first initiating CPB.14 These relatively simple maneuvers can reduce the crystalloid prime to a volume of roughly 200 mL. In addition, technical advances such as leukocyte filters and heparin-bonded circuits reduced the inflammatory response of the body to the blood interface with the CPB circuit, which ultimately can lead to less homologous blood requirement.15,16

The addition of newer pharmacologic and technologic advances to the proven “core” conservation measures as established by Cosgrove and Ovrum had the potential to markedly reduce and even eliminate the need for transfusion even in the face of the more difficult patient characteristics increasingly being encountered. This then became the goal of blood conservation programs.

Identification of Patients at Risk

The coagulopathy associated with CPB is related primarily to the interaction of blood components with the artificial surfaces of the CPB circuit, which results in derangements in platelet function, abnormal functioning of the coagulation cascades, and excessive fibrinolysis. The administration of high-dose heparin to prevent coagulation within the CPB circuit and hypothermia achieved during bypass further contribute to hemostatic derangements. Finally, although the use of asanguineous crystalloid prime rather than the whole-blood pump prime used historically has reduced the amount of blood transfused during CPB dramatically, the resulting hemodilution contributes to the risk of low intraoperative and postoperative hematocrit, which can be independent risk factors for transfusion in the postoperative period. Other risk factors can be assessed in the preoperative state that can identify those patients who may be at high risk of bleeding or who have a low red cell mass, both of which may require autologous blood transfusions.

One of the most important predictors of postoperative bleeding in the surgical patient is a personal or family history of any excessive bleeding or documented bleeding disorders. Many disorders can be confirmed with simple laboratory tests demonstrating some level of coagulation derangement. However, in the cardiothoracic patient population, medications and acquired medical diseases and their associated hemostatic defects are likely to be the most common risk for bleeding. Notably, the use of aspirin alone or included in other medications intended for pain relief or treatment of other ailments is very common. The prevalence among patients undergoing unplanned surgery may be as high as 50%, and this may be even higher among patients with previously diagnosed coronary artery disease.17–19 The currently published data suggest that this does not represent a significant bleeding risk, and there is little evidence to suggest that bleeding times correlate with operative blood loss in these patients.20 Heparin will inhibit factors II and X, and may lead to immune-mediated thrombocytopenia. Coumadin can block gamma-carboxylation and lead to multiple factor deficiency.

Herbal Extracts for Cardiovascular Health

The use of herbal extracts and complementary medicine has become popular in the prevention of arterial thrombotic disease. In 1997, an estimated $21 billion was spent in the United States on complementary and alternative therapies.21,22 Herbs such as thyme and rosemary have been shown to have a direct inhibitory effect on platelets.23

Despite the increased potassium contained in fruits and nuts, menu planning for patients on warfarin can include a healthy diet of these foods without compromising the stability of their oral anticoagulation therapy because most fruits are not important sources of vitamin K, with the exception of some berries, green fruits, and prunes.24 In a case report regarding fish oil supplementation, it was demonstrated that additional anticoagulation could have resulted from an interaction with warfarin therapy. This case reveals that a significant rise in the international normalization ratio (INR) occurred after the dose of concomitant fish oil was doubled. Fish oil, an omega-3 polyunsaturated fatty acid, consists of eicosapentaenoic acid and docosahexaenoic acid. This fatty acid may affect platelet aggregation and/or vitamin K–dependent coagulation factors. Omega-3 fatty acids may lower thromboxane A2 supplies within the platelet, as well as decrease factor VII levels.25

The popularity of herbal additives can be evidenced by the more than $700 million spent annually and the continued expected spending on such products.26 Despite the potential benefits of many herbs with regard to improved well-being, many adverse risks exist. Primarily there is an increased cardiovascular risk among the more commonly used supplements, herbs such as garlic, ginger, and gingko are associated with platelet dysfunction–derived bleeding, whereas supplements such as ginseng and licorice may lead to elevated blood pressure.27

Health-care workers can play a crucial role in identifying possible drug interactions by asking patients taking warfarin about herbal and other alternative medicine product use. Furthermore, the clinical importance of herb–drug interactions depends on many factors associated with the particular herb, drug, and patient. Herbs should be labeled appropriately to alert consumers to potential interactions when used concomitantly with drugs and should recommend a consultation with one's general practitioner.

Autologous Blood Donation

One of the primary concerns in blood conservation is the patient's size and preoperative red blood cell volume. Two early reports by Cosgrove and Utley demonstrated these factors in addition to preoperative anemia as independent risk factors for blood transfusion.28,29 As discussed later in this chapter, there are several relatively simple manipulations to the CPB circuit that can be made to reduce the amount of hemodilution that the patient experiences while on bypass to reduce the overall risk of transfusions.

Preoperative autologous donation (PAD) is a recognized strategy to reduce the risk of homologous blood transfusion in the perioperative period. Although this technique has been in practice since the 1960s, its use in cardiac surgery did not achieve widespread acceptance until the 1980s, with the advent of HIV, as an effort to reduce homologous blood exposures. Unfortunately, in cardiac surgery, the acuteness of the operations and dealing with an older and sicker patient population often preclude PAD because there must be enough preoperative time for autologous collection as well as red blood cell mass regeneration before arriving in the operating room.

Several preoperative characteristics can identify the cardiac patient who is eligible for PAD. The first criterion is that the patient be able to wait the required time for donation and red blood cell regeneration. This length of time typically varies depending on the type of surgical procedure planned (larger operative procedures likely requiring a larger amount of blood) and patient characteristics (eg, body size, blood volume, and hematocrit). In general, this time is a minimum of 2 weeks per unit of blood donated to allow for red blood cell regeneration. The second criterion is that the patient be healthy enough to undergo donation. This criterion would preclude patients with severe left main stem stenosis, critical aortic stenosis, congestive heart failure, and idiopathic hypertrophic subaortic stenosis, as well as patients with severe coronary artery disease and ongoing ischemia, given that many of these patients would have been screening failures for the first criterion. The third criterion is that the patients not have active endocarditis. The time between donation and receiving the PAD unit is ample time for bacteria to replicate in the donated unit with resulting bacteremia, which is potentially life threatening. The fourth criterion is that the patient has an adequate hematocrit and red blood cell mass. A preoperative hematocrit of less than 33% regardless of red blood cell mass is a contraindication to PAD according to the American Association of Blood Banks (AABB) guidelines. A patient with a hematocrit of greater than 33% may be eligible provided that other criteria are met.

Several options exist for patients who historically were not eligible for PAD. Recombinant erythropoietin can be used to accelerate red blood cell production in anemic patients; this strategy is used commonly in Jehovah's Witness patients to increase their red blood cell mass before surgery.30 However, this can be quite costly; as a result, it is usually reserved for patients who are unable to tolerate homologous blood transfusions whether for religious reasons or because they have a rare blood type. An alternative strategy is to stimulate the body to increase the release of endogenous erythropoietin by allowing patients with hematocrits below the traditional cutoff to undergo PAD. The resulting anemia experienced by the patient in the postdonation period is a strong stimulant for endogenous erythropoietin production, ultimately leading to an increase in red blood cell mass.30

Red blood cell mass is related to patient body size.31 A traditional cutoff of 110 lb had been used for PAD. However, the AABB does make specific allowances for PAD in smaller patients. The current recommendation is that no more than 15% of the patient's effective blood volume should be removed at any given time, which takes into account smaller patient body size. PAD should be pursued aggressively for these small patients with low red blood cell mass and hematocrit because they are at highest risk of receiving a blood transfusion at some time during their hospital stay. The mean rate of red blood cell generation of the studies that provided adequate data is 0.46 units per week, or slightly less than 1 unit every 2 weeks.32 In conjunction with PAD, oral iron therapy should be initiated at the time of first donation to ensure adequate iron stores for red blood cell regeneration.

Owing to the relatively acute illness of the population, rarely is there sufficient time for PAD in the cardiothoracic surgical patient.33 In addition, PAD has been supplanted largely by intraoperative blood salvage techniques for reasons of cost-effectiveness (ie, blood withdrawal, preparation, storage, and potential erythropoietin therapy add to costs) and advances in intraoperative blood salvage techniques such as intraoperative autologous donation (IAD; discussed later in this chapter), retrograde autologous prime of the CPB circuit, use of cell salvage, and regular use of cardiotomy suction.

A number of drugs have been used to decrease blood loss and the use of blood transfusion associated with cardiac surgery. Interest has been renewed recently in antifibrinolytics, a relatively old class of drug. Currently, three such medications are used clinically; two are synthetic antifibrinolytics (ie, epsilon-aminocaproic acid and tranexamic acid), and one is naturally occurring (ie, aprotinin derived from bovine lung). Linked with the resurgence of interest in these drugs is interest in diminishing homologous blood transfusions related to cardiac surgery.

Epsilon-Aminocaproic Acid

Epsilon-aminocaproic acid (EACA) is a synthetic antifibrinolytic agent first described in 1959 that derives its effect by forming reversible complexes with either plasminogen or plasmin, saturating the lysine-binding sites and thus displacing plasminogen and therefore plasmin from the surface of fibrin. This blockage of plasminogen binding to fibrin blocks plasminogen activation and therefore fibrinolysis. The overall effect is to block the dissolution of the fibrin clot.

Several studies have demonstrated the efficacy of EACA in reducing bleeding after open heart surgery. Although EACA was first used in situations in which excessive bleeding was encountered, Lambert and colleagues used EACA in 1979 to successfully treat patients with coagulation disorders, who comprise 20% of total primary coronary bypass patients.34 Del Rossi used EACA in 1989 as a prophylactic tool to reduce blood loss in 350 patients undergoing CPB, having an impact on both blood loss and transfusion of autologous blood and blood products.35

Tranexamic Acid

The mechanism of action of tranexamic acid (TA) is similar to that of EACA. The significant difference between EACA and TA is that TA is roughly 10 times more potent than EACA. As with EACA, postoperative bleeding and homologous blood requirements were similarly decreased with the use of TA, as shown by Horrow in 1990 in 38 patients undergoing CPB.36

Aprotinin

Aprotinin is a serine protease inhibitor with antithrombotic, antifibrinolytic, and anti-inflammatory effects. It was recently removed from the market. We include it for historical perspective. It was effective in reducing bleeding and the need for blood transfusions after cardiac surgery with cardiopulmonary bypass. Additional benefits, such as cerebral protection, were hypothesized but not yet thoroughly substantiated.

Aprotinin is a low-molecular-weight serine-protease inhibitor with a lysine residue occupying its active center. It is a naturally occurring polypeptide isolated from bovine lung. Reversible enzyme–inhibitor complexes with various proteases are formed that display activity against trypsin, plasmin, streptokinase–plasma complex, tissue kallikrein, and plasma kallikrein. Because aprotinin is a nonspecific serine antiprotease, it intervenes in the coagulation cascade in multiple loci, working to decrease bleeding in CPB patients by its antiplasmin and antikallikrein effects.

Aprotinin was first used in cardiac surgery by Tice and colleagues, who described the administration of 10,000 to 20,000 kallikrein inhibitory units (KIU), resulting in rapid establishment of hemostasis in five patients who had undergone CPB; they also demonstrated increased bleeding and increased fibrinolytic activity.37 The currently accepted regimen of high-dose aprotinin (5 to 6 million unit average total dose) is reflective of the experiences of the Hammersmith group, which serendipitously saw a reduction in bleeding in patients receiving aprotinin to reduce kallikrein-mediated lung inflammation during CPB.38 Since this report, multiple reports have emerged citing the efficacy of aprotinin as compared with other antifibrinolytics and controls in reducing bleeding complications, homologous blood transfusions, and inflammation primarily in the reoperative setting.39–41 Despite these advantages, Mangano and colleagues reported that aprotinin use was associated with an increased risk of renal failure, myocardial infarction, and stroke.42 In this nonrandomized observational trial of 4374 patients undergoing revascularization, the group concluded that aprotinin use was not prudent in comparison with less expensive alternatives such as aminocaproic acid and TA. Limitations to this study include nonblinded randomization and a significantly higher operative risk in the aprotinin study group that required complicated propensity adjustment to statistically match groups.

In October 2007, the Blood Conservation Using Antifibrinolytics in a Randomized Trial (BART) study was stopped prematurely after preliminary data from 2163 patients demonstrated a trend toward an increase in all-cause 30-day mortality in patients receiving aprotinin. Based on this preliminary finding, the marketing and distribution of aprotinin was suspended.43

Recombinant Activated Factor VII

The use of recombinant factor VII has been shown to induce hemostasis in patients with severe hemophilia A or B with factor specific inhibitors.44 Recombinant activated factor VII complexes with all available tissue factor to activate factor X directly and induce thrombin generation. This leads to the formation of a tight and stable fibrin plug that is resistant to early fibrinolysis.45

Off-label use in cardiac surgery has been indicated when all efforts at conventional hemostasis have been exhausted. These include non–red blood cell product support (eg, fresh-frozen plasma, platelets, and cryoprecipitate), topical agents, desmopressin, and antifibrinolytics. When hemostasis is still not satisfactory, we have given a recombinant factor VIIa (rFVIIa) dose of between 75 and 100 μg/kg. Efficacy is variable with rFVIIa use, ranging from no difference from standard therapy to causing mortal thrombosis. Bruckner reported evidence of thrombosis during support with left ventricular assist when giving high doses for patients requiring higher doses (30 to 70 μg/kg) had a dramatically higher incidence of serious thromboembolic events.46

Heymann and colleagues reported in a retrospective analysis that rFVIIa was safe with regard to risk of thrombosis, but blood loss and transfusion rates were similar. High mortality rates were seen in both groups in house and at 6 months (approximately 30 and 50%, respectively).47 Conversely, Raivio reported in a retrospective series that rFVIIa was significantly more effective in restoring hemostasis; however, severe postoperative thromboembolic complications occurred in 25% of the patients.48 The rFVIIa doses in these two studies ranged from 60 to 80 μg/kg and from 24 to 192 μg/kg, respectively.

Diprose and colleagues compared rFVIIa to placebo in a randomized, double-blind study following complex non–coronary artery cardiac surgery. The group found in this 20-patient study that homologous transfusions were significantly fewer in the study group cohort. There were no differences in regard to adverse events, but this pilot study had limitations that include being underpowered and prone to type I error.49 Despite the effectiveness that has been demonstrated in single-patient case reports and small-scale studies, we recommend caution when using this product until more prospective, randomized studies are available to determine safety and effectiveness.

The limited efficacy of topical agents such as oxidized cellulose and microfibrillar collagen has led to the development of new products with novel applicator systems that are direct activators of the clotting cascade. Additionally, pericardial lavage with aprotinin or EACA has been shown to be ineffective at reducing transfusion requirements and may enhance mediastinal adhesion formation.50–52 As a result of the limitations of more traditional agents, the new-generation topical agents have gained appeal owing to their inherent effectiveness as well as their ease of use.

BioGlue (CryoLife, Inc., Kennesaw, GA) is a biologic glue approved initially for use in the repair of aortic dissection. This product is composed of purified bovine serum, albumin, and glutaraldehyde. The action is almost instantaneous because glutaraldehyde exposure leads to the tenuous binding of lysine molecules, proteins, and tissue surfaces. Raanani described the use of BioGlue as an aid in aortic reconstructive surgery, avoiding the use of stiff Teflon felt strips.53

Tisseel VH (Baxter Healthcare Corp., Glendale, CA) is a topical protein solution sealer that is sprayed onto hemorrhagic surfaces. This fibrin sealant contains fibrinogen, thrombin, calcium chloride, and aprotinin. When the protein and thrombin solutions are mixed and sprayed topically, a viscous solution that sets rapidly into an elastic coagulum is produced. A study by Rousou demonstrated that fibrin sealant was safe with regard to viral transmission and highly effective in controlling localized bleeding in cardiac operations.54

The gelatin-based hemostatic sealant FloSeal (Fusion Medical Technologies, Inc., Mountain View, CA) activates the clotting cascade and simultaneously forms a nondisplacing hemostatic plug. A FloSeal kit contains a bovine-derived gelatin matrix, a bovine-derived thrombin component, and a syringe applicator. The gelatinous matrix is biocompatible and reabsorbed in 6 to 8 weeks. In a series of patients undergoing open heart procedures, the Fusion Matrix Group studied FloSeal versus Gelfoam-Thrombin in procedures in which standard surgical means were ineffective at controlling bleeding.55 FloSeal stopped bleeding in a significantly higher number of patients than did standard therapy with no differences in adverse events; however, there was no mention of each product's effect on transfusion rates.

The two groups of platelet inhibitors are GP IIb/IIIa blocking agents and the adenosine diphosphate (ADP) receptor site blockers. The first group acts via inhibition with the glycoprotein (GP) IIb/IIIa receptor antagonists, which has greatly reduced the need for emergent CABG in patients undergoing angioplasty or coronary stenting procedures. Other adjuncts include the adenosine diphosphate (ADP) receptor blockers. These agents pose new challenges in terms of bleeding risk for the cardiac surgical team for patients exposed to these agents who are subsequently referred for surgery.

GP IIb/IIIa Inhibitors

The final common pathway of platelet aggregation leading to coronary artery occlusion is the cross-linking of receptor GP IIb/IIIa on adjacent platelets by adhesive plasma proteins (fibrinogen). As a result, receptor blockade of GP IIb/IIIa results in blocking platelet aggregation and subsequent thrombus formation. A population of patients with ischemic heart disease who may be expected to require CABG despite maximal therapy with GP IIb/IIIa inhibitors still remains. Currently, there are no controlled studies of the effect of GP IIb/IIIa inhibitors on patients undergoing CABG. However, the use of these inhibitors may enhance the risk of bleeding compared with elective procedures in patients not receiving these agents. The currently used agents are summarized in Table 13-1.

Guidelines for GP IIb/IIIa Inhibitor Patients Requiring Cardiac Surgery

1. Stop the GP IIb/IIIa inhibitor.

2. Delay surgery (if possible, if the patient remains stable) for up to 12 hours if abciximab, tirofiban, or eptifibatide is used, and up to 7 days if clopidogrel is used.

3. Maintain standard heparin dosing despite elevated bleeding times.

4. Use ultrafiltration via the zero-balance technique while on CPB.

5. Transfuse platelets as needed as opposed to prophylactically, preferably once off CPB and after protamine administration.56

Guidelines for Adenosine Diphosphate (ADP) Inhibitor Patients Requiring Cardiac Surgery

Adenosine Diphosphate (Adp) Inhibition

Clopidogrel has become an essential component of therapy in patients with acute coronary syndrome because it significantly improves patient outcomes.

However, there are drawbacks to this therapy, which include delayed onset of action, large interindividual variability in platelet response (as high as 30% nonresponders), and irreversibility of its inhibitory effect on platelets.57–59 Some of the new ADP inhibitors have the advantages of reversible receptor blockade as well as shorter half-lives, which may offer the cardiac surgical team the luxury of offering shorter waiting times with potentially less bleeding complications and ultimately safer operations for the patients who are referred for surgery.

Several companies have developed assays to assess platelet function (platelet aggregation) in the face of using these platelet inhibitors. It is our recommendation to use these studies because they may be cost effective ways of determining the patient's readiness for surgery as a routine part of the standard coagulation profile.

Heparin-Induced Thrombocytopenia and Cardiac Surgery

Heparin-induced thrombocytopenia and thrombosis (HITT) is a serious complication of heparin therapy. This immunologic response develops in approximately 1 to 3% of patients treated with unfractionated heparin (UFH) for 5 to 10 days.60,61 HITT is a severe prothrombotic condition, with affected individuals having a 20 to 50% risk of developing new thromboembolic events. The mortality rate is approximately 20%; in addition, approximately 10% of patients require amputations or suffer other major morbidity.62,63 Although there is no best way to anticoagulate these patients, when surgical delay is unavoidable, few options exist. Most reports, however, are limited single case reports with significant morbidity and mortality.

Currently, direct thrombin inhibitors (DTIs) are recommended for HITT (Table 13-2). Limitations of these agents include the lack of a reversal agent and route of clearance.64

These agents have several advantages over UFH. Because they do not rely on antithrombin III levels and do not bind to plasma proteins, the anticoagulant effect is more predictable. Also, DTIs inhibit fibrin-bound thrombin as well as fluid-phase thrombin, leading to a greater antithrombotic effect.

Specific recommendations on how to treat patients with HITT cannot be given at this time. Diagnosis of HITT is difficult because assays vary in both specificity and sensitivity. Second, and vital to this dilemma, are the lack of randomized trials. Most data available for treating these patients are derived from single case reports and small retrospective trials. Dosing of the available anticoagulation agents is not fully understood, and there is no reversal agent. Because drug elimination varies for these agents, renal or liver function must be fully assessed.

Intraoperative Period

Intraoperative autologous donation (IAD) of whole blood has many advantages over PAD. IAD does not require a delay in surgery; it can be performed efficiently and with minimal additional cost. In addition, the blood product obtained by IAD is whole blood, which is transfused within 2 to 3 hours of collection and therefore contains active platelets and factors. The resulting advantages are the avoidance of coagulopathy frequently seen after CPB and the addition of red blood cell mass capable of oxygen transport. The storage process limits the amount of blood loss during surgery via lap pads and discard suction. Additionally this technique spares the damaging effects of the heart-lung machine, which include contact activation of platelets and complement as well as red blood cell hemolysis. Because IAD serves to decrease red blood cell requirements in a volume-dependent manner, the maximum amount of blood should be removed from each individual patient so as to optimize blood conservation efforts. The amount of blood that an individual patient is capable of donating via IAD depends strictly on the patient's own physiologic parameters, estimated blood volume (based on height-weight nomogram), pre-IAD hematocrit, and pre-CPB hematocrit.65 In Fig. 13-1 is our nomogram for allowable IAD blood drainage. It conservatively estimates the volume of blood that can be removed to achieve a hematocrit of 24% or greater (based on a 1000-mL CPB prime volume).

Minimal hemodilution by red blood cell priming of the CPB circuit with the patient's own blood is a useful technique in avoiding transfusions. Techniques include low-prime circuitry and retrograde autologous prime (RAP).66,67 Up to 90% of the crystalloid prime of the CPB circuit can be displaced with autologous blood using RAP. This technique involves partial priming of the bypass circuit with the patient's own blood from the arterial cannula just before instituting CPB. In addition, the venous loop also can be primed in a similar manner when first instituting CPB, displacing the crystalloid prime in the venous loop. The result is the replacement of virtually all the crystalloid prime with autologous blood, thereby reducing the amount of hemodilution the patient experiences on CPB. Available low prime circuitry (<1000 mL) allows for complete autologous priming of the CPB circuit and less hemodynamic alteration during RAP.

The literature indicates that the anesthetized patient on full CPB at moderate hypothermia can safely tolerate a hematocrit as low as 15%, with the exception of patients at risk for decreased cerebral oxygen delivery, namely, those with a history of cerebrovascular accident (CVA), diabetes, or cerebrovascular disease.68 These latter patients can tolerate a hematocrit as low as 18% when using moderate hypothermia.69 Once the patient is warm and being weaned from CPB, these percentage points are raised by 2% each (17 and 20%, respectively) because the relative protective effects of the hypothermia are no longer present. In our institution, once the patient is off CPB, our practice has been to retransfuse all or as much as possible remaining blood in the CPB circuit to the patient and then give all available cell salvage blood, including any blood remaining in the CPB circuit that was not initially given back to the patient, then any IAD blood, and then finally PAD blood if available. Then and only then, if the hematocrit remains unsatisfactorily low, does the patient receive homologous blood.

Once the patient leaves the operating room, we use a transfusion trigger corresponding to a hematocrit of 22% in asymptomatic patients. In patients older than 80 years of age, a trigger of 24% is used. These numbers are meant to serve as guidelines; if a patient is at all symptomatic (ie, tachycardic, hypotensive, ischemic, or showing any evidence of end-organ hypoperfusion), he or she will receive homologous blood transfusion therapy. To apply minimum transfusion standards safely and appropriately, the cardiac surgeon or anesthesiologist must have an understanding of the lowest safe level of anemia under the variety of conditions encountered by the cardiac surgical patient. This understanding then can be combined with an assessment of the patient's clinical status to determine the true need for red blood cell transfusion.

Blood conservation is an important component to improving health care. This process has the ability to improve patient outcome, reduce cost, and increase the availability of such an important commodity.70 The reduction of transfusions has been shown in many studies to improve surgical results and even eliminate certain complications.71–73 As cardiac surgery patients present with higher acuity it becomes ever more challenging to reduce exposure to banked blood products. Arora demonstrated that patients with a preoperative hemoglobin of 12 g/dL or less had a significant risk for transfusion.78 Other risk factors for transfusion most notably were emergent procedure and renal failure at time of surgery.

Our group reviewed the use of heparin-coated CPB circuits and the effect on blood usage and pulmonary function in 200 patients. As seen in Table 13-3 coated circuitry reduced overall blood usage and resulted in reduced ventilator time and time to ambulation. An interesting finding was that patients who were not transfused postoperatively had improved outcomes as well, regardless of whether they had coated circuitry or not. One explanation is that if blood transfusion occurs on bypass, the lungs are not immediately exposed to the transfused blood because the homologous blood is given into the arterial rather than the venous system, thus avoiding a first-pass effect. Typically if a transfusion is given postoperatively it is infused into a central vein, thereby having almost immediate exposure to the lungs. Thus the timing of blood transfusion may be important.

If we review the “two-hit” lung injury model of transfusion-related acute lung injury (TRALI), transfusion immediately post-CPB may be a similar scenario.74,75 The “first hit” is usually described as an underlying illness that primes the pulmonary endothelial cells and leukocytes or an inflammatory response incited by an event such as cardiopulmonary bypass itself. The “second hit” is delivered by the transfusion of red blood cells into the venous system. If we look at this model of TRALI, the optimal timing of blood transfusion for the typical patient undergoing a procedure utilizing cardiopulmonary bypass may be before separating from bypass to avoid the first-pass effect and thereby the “second hit.” Further studies will be required to substantiate this assumption, however.

Intuitively the smaller patient with a low preoperative hematocrit is at high risk of transfusion. We reviewed our data related to this and found that all patients less than or equal to 70 kg were transfused during their length of stay (Table 13-4). These patients also had the highest incidence of respiratory failure and the need for rehabilitation therapy. This has led us to change our transfusion practice whereby we attempt to transfuse all red cells on cardiopulmonary bypass. Initially we started this practice for patients less than or equal to 70 kg with hematocrits of 28% or less. With the use of lower prime perfusion circuitry (<1000 mL) we have reduce the threshold to 25% or less for when we will blood prime. These values and thresholds will change as we continue to optimize the perfusion circuitry.

There is no question that some aspects of the blood conservation approach are more costly than others. However, when taken in the context of the cost of blood products, their use, the potential reduction in the risk of reexplorations, and the reduction or elimination of the risks of homologous blood exposures, these costs may seem a little more reasonable. Overutilization of blood products leads to increased direct and indirect costs. The cost for a unit of red blood cells is over $1400.76 This includes approximately $600 for the direct product cost and more than $800 for the adverse event cost related to transfusion. Other implications of transfusion include increased mortality and morbidity when “older blood” that is greater than 14 days old is transfused.77

As patients present with even higher levels of risk the cardiac surgeon will appreciate any options that may reduce blood exposure. The typical patient is anemic and has many risk factors for significant bleeding because of exposure to platelet inhibition. Fortunately cardiac surgery groups can take advantage of techniques that reduce hemodilution and preserve hematologic function of coagulation. These options include retrograde autologous priming, intraoperative autologous donation, and coated-reduced prime perfusion circuitry.

Historically, cardiac surgery has been associated with a high incidence of blood transfusion, with up to 70% of these patients receiving homologous blood transfusions at some point during their hospital stay. However, with improving technology, awareness of blood conservation techniques, and better pharmacologic agents, a multidisciplinary approach to blood conservation can make “bloodless heart surgery” entirely possible. For the purely elective patient, these techniques are initiated preoperatively with PAD, whereas other patients are eligible for one or all of the remainder of the in-hospital techniques described. Using a team approach that both optimizes and integrates the use of each of these measures, the use of homologous blood can be markedly reduced in a majority of cardiac surgical patients.

• Blood conservation should be approached in a multidisciplinary fashion. This team should include members from departments of surgery, anesthesiology, nursing, perfusion, transfusion therapy, and quality.
• Blood usage data should be collected, reviewed, and discussed at monthly meetings
• Intraoperative autologous donation should be maximized using the blood removal nomogram. This blood product must be labeled and checked as if it were a homologous transfusion.
• The anesthesiologist should minimize hemodilution with judicious maintenance of fluid balance (using vasopressor and fluid).
• The surgeon should minimize blood loss via meticulous technique and minimal lap pads and sponges.
• The perfusionist should minimize hemodilution to negligible levels by optimizing retrograde autologous priming techniques and reduce pump prime.
• The operating room nurses and technicians should be diligent in avoiding blood loss at the operative field and aiding in the availability of blood products when needed.
• Red blood cells should be transfused during cardiopulmonary bypass when needed, rather than waiting until later when they can increase the risk of lung injury.