1. 
   

   What are the indications for transfusion of blood products?

   
   
2. 
   

   What are the common complications of transfusion?

   
   
3. 
   

   What interventions minimize the need for transfusion?

The transfusion of blood products is a regular, daily occurrence in hospital practice. In 2006, in the United States, approximately 22.3 million transfusions of various blood products were carried out. Depending on the size of the hospital and the range of services offered, only a few patients a month or many patients each day may be transfused. Similarly, blood banks may carry as few as one or two to more than a dozen different types of blood products. Table 177-1 shows a list of most of the blood products provided to Canadian hospitals for patient use by Canadian Blood Services.

It is not necessary for physicians outside the field of transfusion medicine to have detailed information about each of these many products. However, clinicians should be able to explain the indications for transfusion of the most commonly used blood products and recognize transfusion reactions. Each of the five cases demonstrates how certain blood products might be used and the adverse outcomes that are potentially associated with their use.

Each of these patients would likely benefit from the transfusion of different products and at least one of them would probably benefit from the transfusion of multiple blood products.

Red Blood Cell Transfusions

The treatment of anemia, with the consequent improvement in tissue oxygen delivery, is the key indication for the transfusion of red blood cells (RBCs). One unit of RBCs has a volume of about 250 mL and should contain a hematocrit of less than 0.80 L/L. Transfusion of one RBC unit will increase the hemoglobin by about 1.0 g/dL (10 g/L) and can often relieve the symptoms in patients with anemia. Treating anemia may also be associated with improved outcomes in certain patient groups. However, there is no hemoglobin level at which a transfusion is indicated for all patients. The preexisting physical condition of an individual, the rate at which the hemoglobin has dropped, as well as the degree of anemia all impact on how well the patient will compensate for the low hemoglobin. The decision to transfuse should be individualized for each patient based on clinical condition. Additionally, the decision should be made with a clear understanding of the risks, as well as the benefits of RBC transfusions.

It is generally accepted that patients with acute blood loss due to active bleeding will need RBC transfusions to maintain the hemoglobin above 7.0 g/dL (70 g/L). Transfusion therapy in such cases should always be considered only one part of a resuscitative measure. Attention should be focused on treating the source of blood loss. In patients with chronic anemia, compensatory mechanisms often have the opportunity to come into play and consequently, patients with chronic anemia may tolerate much lower hemoglobin levels than those with acute blood loss. Indeed, young patients with uncompromised cardiac or pulmonary function may tolerate very low hemoglobin levels. RBC transfusions are usually recommended in older patients or those with poor cardiopulmonary reserve whose hemoglobin levels are less than 7.0 g/dL (70 g/L). Transfusion to higher levels may be necessary in symptomatic patients but transfusion is not usually appropriate if the hemoglobin level is greater than 10 g/dL (100 g/L).

Though the above recommendations are generally accepted, the literature regarding when to transfuse anemic patients presents some conflicting data and can be difficult to interpret. One of the key areas of conflict is the prophylactic transfusion of patients with moderate degrees of anemia (hemoglobin 7.0–10.0 g/dL or 70–100 g/L), especially in patients who are perioperative or critically ill. The best available data regarding the transfusion of critically ill patients come from the Transfusion Requirements in Critical Care (TRICC) trial in which critically ill patients were randomized to either a restrictive or a liberal transfusion protocol and followed prospectively to assess for difference in mortality. Those in the restrictive transfusion group were transfused to maintain their hemoglobin between 7.0–9.0 g/dL (70–90 g/L) and those in the liberal arm were transfused to maintain their hemoglobin between 10.0–12.0 g/dL (100–120 g/L). Mortality at 30 days was similar in the two groups and there was some indication that the restrictive strategy might be superior in younger patients who were less critically ill. This was not the case in the subgroup of patients who experienced myocardial ischemia. These data are usually interpreted to indicate that critically ill patients may safely tolerate a drop in hemoglobin level below 90 g/L before RBC transfusions are indicated, but higher levels may be required in the more severely ill or those with active coronary artery disease.

A large retrospective study examined the effect of perioperative transfusion on mortality at 30 days and 90 days after surgery. No correlation between transfusion and mortality could be found and the results suggested that patients whose hemoglobin levels were as low as 8.0 g/dL (80 g/L) did not suffer increased mortality. However, these conclusions are limited by the fact that nearly all patients in the study with a hemoglobin less than 8.0 g/dL (80 g/L) received transfusions. Despite the large numbers of patients evaluated in the above-noted trials and other studies, there can be no consensus on a threshold hemoglobin value at which all patients should be transfused. It merits frequent repetition that clinical signs and symptoms, not a hemoglobin number, should inform the decision to transfuse RBCS. Most importantly, the potential risks of transfusions should always be weighed against the anticipated benefits.

There are certain special groups of patients whose transfusion requirements differ from that of the general population. For example, patients with sickle cell anemia suffer chronic anemia caused by ongoing hemolysis and often tolerate low levels of hemoglobin. However, when complications arise, transfusion therapy is an important part of their therapy. Two types of RBC transfusions may be indicated in these situations. One is a simple transfusion, the other is an exchange transfusion in which the patient’s RBCs are removed at the same time that donor RBCs are transfused. Simple RBC transfusions are indicated for the relief of the complications of sickle cell disease including, but not limited to, acute anemia due to blood loss, aplastic crisis, acute splenic or hepatic sequestration crisis, and acute chest syndrome. Additionally, there is evidence that such patients may require preoperative transfusion to reach a goal of a hemoglobin level of 10.0 g/dL (100 g/L) if the patient will be undergoing a general anesthetic or if the surgery is ophthalmologic.

When it seems likely that a patient will require an RBC transfusion certain tests should be done in the transfusion medicine laboratory to ensure that a safe product is chosen for the patient. These tests are commonly referred to as the group, screen, and crossmatch. Grouping consists of determining the patient’s ABO and Rh(D) blood groups. It is mandatory that all patients be transfused with RBC units that are ABO compatible (see Table 177-2). It is preferable to transfuse Rh(D) matched units to all patients, but it is especially important to do so in young girls and women of child-bearing age who are Rh(D) negative and therefore at risk of having hemolytic disease of the newborn if they become alloimmunized with anti-Rh(D) antibodies. When blood is needed urgently and grouping cannot be performed, group O Rh(D) negative units, the universal blood donor type, should be used.

Screening of a recipient’s blood sample refers to assessing a patient’s serum for presence of antibodies against RBC antigens. Crossmatching is the final step to ensure compatibility between donor and the recipient. These tests fall into one of two general categories: those tests that involve mixing donor and recipient serum, and those which can be done electronically, known as computer crossmatches. If an actual laboratory test is done, the RBCs of the donor and serum of the recipient are mixed and donor RBCs are evaluated for the presence of agglutination. If no RBC agglutination occurs, then the donor RBCs are compatible for transfusion to the recipient and no further testing is required. If RBC agglutination is seen, further testing on both the donor blood and the recipient’s sample is needed. In an electronic crossmatch, the computer will select an appropriate unit from the blood bank’s supply. The donor unit must have been tested twice to confirm the ABO group and the recipient should have two recent ABO groups and one antibody screen recorded in the hospital’s laboratory information system. The recipient must not have had any transfusions subsequent to these tests being done.

Platelet Transfusions

In general, patients with low platelet counts or documented platelet function abnormalities who are bleeding are likely to benefit from platelet transfusions. However, while all causes of anemia may be treated with transfusion, not all causes of thrombocytopenia may be safely treated with the transfusion of platelet products. Thrombocytopenia caused by bone marrow failure, chemotherapy, bone marrow transplantation, or mechanical peripheral destruction (as occurs with use of cardiopulmonary bypass pumps) may be safely treated with platelet transfusions, whereas conditions associated with immune destruction of platelets (ex ITP or HIT) or microangiopathic hemolysis (ex TTP) can be dangerously aggravated by platelet transfusions.

While decisions to transfuse platelets should always be made on an individual case-by-case basis, the transfusion threshold for platelet transfusions is generally well accepted. The majority of platelet transfusions currently are for the prophylactic treatment of chronic hypoprolifative thrombocytopenia. There is evidence from prospective randomized trials that withholding platelet transfusion until platelet counts drop below 10 × 109/L is a safe practice that enables a rational use of a very limited resource. A higher threshold of 20 × 109/L should be used in patients with additional bleeding risks such as sepsis. Patients requiring surgery or other invasive procedures who are also thrombocytopenic should be treated with prophylactic platelet transfusion. The final decision about what platelet count might be appropriate should be made in conjunction with the physician performing the procedures. Some general recommendations can be made, despite the lack of good evidence-based studies providing guidelines in such circumstances. A platelet count of greater than 100 × 109/L should probably be maintained for neurosurgical procedures and in patients who have suffered head trauma. For other types of procedures associated with an estimated blood loss of greater than 500 mL and those being performed under epidural anesthesia, platelets should be transfused to keep the count above 50 × 109/L. For those procedures associated with lower blood losses, a platelet count between 20–50 × 109/L should usually be sufficient.

The above recommendations are not the subject of much debate; however, questions remain about the dose of platelets that should be given, especially for those patients receiving prophylactic transfusions. A typical dose of platelets contains about 300 × 109 platelets. Studies looking at the safety and efficacy of lower doses versus these standard doses of platelet concentrates for transfusion have provided conflicting results. A recent large clinical study (PLADO) has provided evidence that a platelet dose as low as 110 × 109 is not different from a dose of 440 × 109 in bleeding risk. Until additional high level evidence becomes available, platelet transfusion doses such as those listed above should continue to be prescribed.

Multiple methods exist for the production of platelet products for transfusion. Two general preparation methods are the platelet rich plasma (PRP) method and the buffy coat method. In the PRP method, platelets are separated from a unit of whole blood and suspended in plasma from the same donor. One such platelet unit typically contains 55 × 109 platelets; and five units usually represent an average adult dose. The five units must be pooled in the blood bank just prior to being released for transfusion. In the buffy coat method, platelets from four donors are pooled by the blood collection agency and suspended in plasma from one of these donors. This prepooled unit constitutes one dose and usually contains at least 240 × 109 platelets. The buffy coat method is considered advantageous in that it decreases the number of donors to which the recipient is exposed, more plasma is made available for other uses, and since the units are pooled during production, there is no need to pool them in the hospital, thereby, they are available to leave the blood bank quicker.

Platelet concentrates are also prepared by an apheresis procedure that removes only platelets from the donor’s circulation. Such a unit usually contains approximately 250 × 109 platelets and one unit constitutes one adult dose. While platelets prepared from any of these methods are safe for transfusion, there are some situations in which apheresis platelets may be particularly beneficial. In patients who are likely to be transfused multiple times, apheresis platelets may help reduce the risk of alloimmunization to platelet and WBC antigens by limiting the number of donors to which the recipient is exposed.

The Transfusion of Plasma Products and Plasma Derivatives

The correction of coagulation abnormalities of differing etiologies may be accomplished through the use of plasma and plasma derivatives such as cryoprecipitate and prothrombin complex concentrates. However, not all patients with coagulation abnormalities require transfusion therapy. The 2008 American College of Chest Physicians clinical practice guidelines on the management of anticoagulation recommend using blood products to reverse the anticoagulant effect of oral vitamin K antagonists only if there is evidence of serious bleeding or when urgent surgery is needed. In such situations the use of fresh frozen plasma (FFP) or prothrombin complex concentrates (PCCs) in combination with oral or intravenous vitamin K is suggested. Indeed, whatever the etiology of the coagulation abnormality, only if there is evidence of bleeding or the need for invasive procedures should plasma component or derivative therapy be used.

Immediately following collection, an average unit of FFP contains approximately one clotting factor unit/mL of all the coagulation factors. However, concentrations of the labile factors, V and VIII, decrease rapidly in unfrozen plasma. Consequently, plasma intended for transfusion must be frozen shortly after collection and maintained in a frozen state to ensure that there will be adequate levels of these clotting factors at the time of transfusion. To achieve a level of at least 30% of these clotting factors, a dose of 10–15 ml/kg of FFP is usually needed. The units selected for transfusion must be ABO compatible (see Table 177-3). While group O RBCs are safe for transfusion to patients of all ABO types, this is not the case for plasma. The universal plasma donor is group AB and this ABO type of plasma should be used for emergency transfusions when the recipient ABO group is not known.