Autologous blood transfusion and directed transfusion

Autologous donation before a scheduled surgical procedure and transfusion to the patient during surgery has been shown to decrease allogeneic exposure in routine cardiac and orthopedic surgery, but predonation does not always eliminate the need for allogeneic blood. Predonation of autologous blood is not necessarily less expensive than collection and transfusion of allogeneic blood, nor does it completely eliminate the risks of transfusion reactions.

It is controversial as to whether autotransfusion at the site of care, also known as cell salvage, reduces allogeneic transfusions or reduces costs. Typically, allogeneic transfusion requirements are reduced, but the cost of equipment and personnel far exceeds the cost of collecting and transfusing allogeneic RBCs.

Directed donation is the process in which a patient or patient’s family selects blood that comes from an identified donor, often a relative of the patient. Interestingly, directed donation may be associated with an increased infection risk because the donor who responds to a request to donate for a specific individual is no longer a volunteer; the individual may feel coerced into donating. Directed donation does not eliminate the risk of alloimmunization or immunomodulation, because this blood is allogeneic, and blood from related donors actually increases the risk of graft-versus-host disease.

Synthetic hemoglobins

The use of hemoglobin-based O₂ carriers (HBOCs) has been hampered by difficulty in defining meaningful clinical end points, safety parameters, and risk/benefit ratios. All HBOCs rapidly bind nitric oxide, resulting in increased vascular resistance as well as interference with other functions of nitric oxide. Increased levels of inflammatory cytokines, increased platelet reactivity, and decreased organ blood flow are thought to be responsible for the pancreatitis, esophageal spasm, myocardial injury, pulmonary hypertension, and acute lung injury associated with the use of HBOCs. Reactive O₂ species, resulting from free iron release, may mediate renal and central nervous system injury. A recent meta-analysis found a statistically significant increase in death and myocardial infarction associated with the use of HBOC.

Recombinant hemoglobin-based products have the advantages of O₂-binding characteristics more similar to those of native hemoglobin but are unstable in solution, scavenge nitric oxide, and also release free iron into the bloodstream.

Red blood cell transfusion

The sole reason to transfuse RBCs is to increase the content of O₂ in the blood, thereby increasing O₂ delivery (), which is a product of hemoglobin concentration, arterial O₂ saturation, and cardiac output (CO, which itself is a product of stroke volume and heart rate). A specific hematocrit value may sustain adequate if CO is adequate but may be insufficient when cardiac output is limited or when arterial saturation is impaired by the presence of a transpulmonary shunt. Therefore, despite the widely accepted hemoglobin trigger of 7 g/dL, the decision to transfuse should take into consideration the current hemoglobin level, estimated blood loss, cardiac reserve, vital signs, the likelihood of ongoing hemorrhage, and the risk of tissue ischemia. The dynamic nature of surgical hemorrhage requires a more aggressive approach to blood replacement in the operating room, compared with sites elsewhere in the hospital. In patients with chronic anemia, increased 2,3-DPG levels make O₂ transport (see Chapter 21) more efficient; in acute anemia, cardiovascular mechanisms of compensation (e.g., increased CO, heart rate, myocardial O₂ consumption) are more important.

Indications for transfusion of red blood cells

The binding of O₂ to hemoglobin is represented by a sinusoidal relationship known as the oxyhemoglobin dissociation curve, (see Chapter 21), which facilitates efficient O₂ loading of hemoglobin in the lungs (where PO₂ is high) and unloading of hemoglobin in the tissues (where PO₂ is low). Because the vast majority of O₂ carried in the blood is noncovalently bound to hemoglobin, is the product of CO and O₂ content:

< ?xml:namespace prefix = "mml" />D˙O2=CO×Cao2

where O₂ content is calculated as follows:

Cao2=(1.36 × [Hb]=Sao2)+(0.0031 × Pao2)

Although otherwise healthy patients can make extraordinary adaptations to maintain and consumption in the face of severe anemia, there is evidence that those with cardiovascular and cerebrovascular disease have limited ability to compensate for acute anemia below hemoglobin levels of 7 to 10 g/dL. Myocardial ischemia is often silent and is not always related to the heart rate and blood pressure. Although medical management (β blockade) is important for most of these patients, anemia could add to their risk of infarction. Furthermore, although serial hemoglobin determinations are helpful intraoperatively, they do not reflect acute changes in intravascular volume and can be misleading. Overexpansion of intravascular volume with colloid or crystalloid can produce a lower hemoglobin level in a hypervolemic patient. Alternatively, inadequate administration of crystalloids or excessive diuresis can lead to a normal or high hemoglobin level in a hypovolemic patient.

A study that examined outcomes in adult patients in the intensive care unit who were randomly assigned to a restrictive transfusion strategy (target hemoglobin 7 g/dL) or a liberal strategy (target hemoglobin 9 g/dL) found no difference in overall mortality rate between the groups and a lower in-hospital mortality rate for the restrictive group in a subgroup of patients with lower APACHE (Acute Physiology, Age, and Chronic Health Evaluation) scores. In addition, myocardial infarction and pulmonary edema were also more frequent in the liberal-strategy group. In patients with coronary disease, transfusion was not associated with improved survival nor were mechanically ventilated patients more likely to be weaned from the ventilator. Similar results have been confirmed in preterm infants.

Platelet matching

Platelets are not routinely matched for ABO compatibility because expression of the A and B antigens on platelets is believed to be of either minimal or no significance. However, recent evidence indicates that ABO-incompatible platelet transfusions have decreased efficacy, as compared with those that are ABO-matched. Although platelets do not express Rh antigens, platelet transfusions are matched for Rh compatibility. That is, Rh-positive platelets are administered only to Rh-positive recipients because a small number of RBCs are almost invariably present in platelet concentrates and could, theoretically, alloimmunize an Rh-negative recipient. Despite this theoretical concern, recent studies of Rh-incompatible platelet transfusions have shown that this is probably not a significant risk.

Indications for platelet transfusion

Box 102-1 summarizes the indications for platelet concentrate transfusion listed in the American Society of Anesthesiologists’ Guidelines for Perioperative Transfusion Therapy. Patients with abnormal platelet function or thrombocytopenia are likely to benefit from administration of platelet transfusions if the platelet disorder is thought to induce or exacerbate their bleeding. Platelet counts of less than 10 × 10⁹/L often occur in patients receiving chemotherapeutic agents. Platelet transfusions are used to prevent spontaneous intracranial and gastrointestinal hemorrhages in these patients.

For major surgical procedures in thrombocytopenic patients, it is desirable to increase the platelet count to 50 × 10⁹/L to 100 × 10⁹/L, and prophylactic administration of platelet transfusions is indicated. Platelet transfusion is not indicated simply to increase platelet counts in patients who are neither bleeding nor about to undergo interventional procedures. Patients with immune thrombocytopenic purpura should not receive platelet transfusions unless they have life-threatening bleeding. These patients produce autoantibodies that react against all human platelets, and, thus, they derive little to no benefit from a platelet transfusion.

Following cardiopulmonary bypass, most patients develop both thrombocytopenia and a functional platelet impairment. Although the correlation between platelet counts and the extent of bleeding in these patients is poor, transfusing based on algorithms using platelet count as an indication for platelet transfusion reduces the number of platelets actually used.

Functional platelet disorders are encountered less frequently than is thrombocytopenia. In addition to cardiopulmonary bypass, uremia, liver disease, myeloproliferative disorders, and dysproteinemias can cause acquired functional platelet disorders. Drugs that affect cyclooxygenase (aspirin, nonsteroidal anti-inflammatory drugs), theophyllines, tricyclic antidepressants, anesthetic agents (especially halothane), and some antibiotics cause functional platelet disorders that may or may not become clinically significant. Inherited functional platelet disorders include Glanzmann thrombasthenia, Bernard-Soulier syndrome, gray platelet syndrome, and dense granule deficiency syndrome.

Platelets have dozens of known proteins on their surfaces, and polymorphic variants have been identified in almost all of these proteins. Platelets also express HLA antigens. As a result, platelets from nonidentical donors are antigenic, and 24 immunologic platelet-specific antigens have been defined serologically. Sensitization to platelet antigens is common in patients who have received multiple platelet transfusions. Patients who are sensitized to these antigens or to HLA antigens will rapidly destroy transfused platelets, decreasing the therapeutic effectiveness of the platelet transfusion. In sensitized patients, only type-specific matched platelets are effective. Leukodepletion has been shown to be effective in reducing the incidence of platelet alloimmunization.