Keywords

FormalPara IFA Commentary (MLNGM)

Blood transfusions are an integral part of treatment in the critical care setting. Like any other medical interventions, transfusions are also associated with serious adverse reactions which may affect patient outcomes. Evidence from various RCTs show that adhering to a restrictive RBC transfusion strategy is more beneficial than liberal transfusion strategies. The age of the transfused RBCs shows minimal to no effect on patient outcomes. Similarly, plasma and platelet components are also recommended by experts and guidelines for therapeutic purposes or prophylactically prior to any intervention or prevention of life-threatening bleeding. Point-of-care tests such as thromboelastography (TEG) or rotational thromboelastometry (ROTEM) are useful aids in guiding transfusion therapy. Use of transfusion alternatives such as erythropoietin and tranexamic acid reduce the overall need for blood transfusions and hence the associated adverse effect.

In hemorrhagic shock, the macrocirculation, microcirculation, and tissue perfusion are impaired due to massive blood loss. Infusion strategy aims at restoring the volume of the intravascular space and, thereby, facilitating oxygen transport. The clinical questions include indication and differential indication of crystalloid and colloid fluids, clinical targets, and dosing. Intravenous fluids are required in order to restore microcirculation and to prevent organ dysfunction and death in massive bleeding. Deliberate hypotension is recommended in uncontrolled hemorrhage. In uncontrolled hemorrhage in general, pre-warmed fluids should be administered in order to prevent hypothermia-dependent coagulation disturbance. In hemorrhagic shock with lactate acidosis, additional hyperchloremic acidosis due to saline-based infusions should be avoided by the use of chlorine-balanced solutions.

Various crystalloid and colloid solutions are available. In a theoretical pathophysiology-driven approach, crystalloids are indicated to replace extravascular deficits and colloids are indicated for intravascular volume replacement. The use of crystalloids only cannot fulfill a pathophysiology-driven fluid strategy because of high volume loss into the interstitial compartment. Historically, in countries with predominant crystalloid resuscitation, coagulation management was based on a 1:1:1 ratio concept of red blood cell concentrates vs. fresh frozen plasma (FFP) vs. platelet concentrates. Ratio-based transfusion regimens deliver relevant amounts of volume (three components together about 600 ml). The volume expanding capacity of the 8.5% protein solution in FFP, however, is unknown. Albumin is used in some countries as an endogenous colloidal solution in massive bleeding but both FFP and albumin also have their disadvantages, risks, and costs.

Coagulation factor concentrate-based coagulation management delivers procoagulant activity in small carrier solutions (50 ml), and synthetic colloids with a context-sensitive volume expanding effect of around 100% are often used in this setting. Synthetic colloids, however, may aggravate bleeding by inducing intravascular dilutional coagulopathy. Accordingly, maximum doses need to be considered. In acute bleeding, the endothelial barrier is suggested to be intact.

Fluid strategy in hemorrhagic shock is heterogeneous (from liberal to restrictive) throughout countries and continents. Studies comparing traditional regimens are warranted. In a pathophysiology-based concept, balanced crystalloids plus colloids are given individualized according to metabolic (and preload) parameters with monitoring for (dilutional) coagulopathy and active avoidance of overdosing and hypervolemia.

FormalPara Learning Objectives

The learning objectives of this chapter are:

  1. 1.

    To list the indications for red blood cell (RBC) transfusions in various subgroups of critically ill patients.

  2. 2.

    To assess the effect of the age of RBC on patient outcomes.

  3. 3.

    To review the indications for platelet transfusion in critically ill patients.

  4. 4.

    To explain the role of plasma transfusion in coagulopathic, critically ill patients.

  5. 5.

    To list the frequently encountered and serious transfusion reactions in ICU.

  6. 6.

    To discuss alternatives to blood transfusions.

FormalPara Case Vignette

Mr. G, a 34-year-old male, presented to the emergency department with a road traffic injury. On arrival, he was conscious but agitated with heart rate of 150 beats per min, respiratory rate of 30 breaths per minute, and blood pressure of 80/50 mmHg. He was immediately put on oxygen. Two large-bore IV lines were placed and a liter of crystalloid solution was started. Massive transfusion protocol was initiated. Further examination revealed pelvic fracture and hemoperitoneum. He underwent exploratory laparotomy, which revealed splenic laceration and retroperitoneal hematoma. Splenectomy was performed with approximately 3 liters of blood loss. The patient was transferred to ICU with vasopressor support and ongoing packed red blood cell (PRBC) transfusion.

Questions

  • Q1. What are the indications for blood and its components in ICU?

  • Q2. Does the age of RBCs matter?

  • Q3. What is the management of coagulopathy in critical illness?

  • Q4. What are the adverse effects of blood transfusion?

Introduction

Blood transfusions are administered in over half of all patients admitted to ICU. The decision to transfuse blood and blood components in the ICU setting is often determined by present need and the underlying comorbidities in the patient. However, it is now more evident that adverse consequences of blood transfusions are common in ICU. Thus, it is necessary to balance the risks and the benefits, before the decision is made. Evidence-based practice in the ICU can provide valuable guidance in these situations. In this chapter, we discuss the available evidence for blood transfusion in various subgroups of critically ill patients. We will also address other common issues pertaining to the transfusion of blood and blood components that can affect the outcomes in these patients.

Anemia and Red Cell Administration

It is estimated that approximately 25% of patients admitted to the ICU are anemic at the time of presentation, whereas nearly 40% of patients become anemic (Hb <9 g/dl) during the course of their ICU stay [1, 2]. Anemia in critical illness results from factors such as blood loss, hemodilution, and decreased production and/or increased destruction of red blood cells (RBC). In these patients, RBC transfusion is most commonly performed for immediate correction of anemia with the aim to improve oxygen-carrying capacity and thus tissue oxygenation. However, RBCs are also transfused to promote hemostasis through its rheological effect. This effects allows the RBCs to preferentially move towards the center of blood vessel, causing margination of platelets and plasma, ensuring improved hemostasis [3].

Traditionally, the trigger for RBC transfusion is Hb <10 g/dL or hematocrit level < 30%, known as the 10/30 rule. Although RBC transfusions can often be lifesaving in the critically ill, there is an increasing body of evidence that it is associated with increased risk of infection, hospital stay, acute lung injury, multiple-organ failure, and mortality in a dose-dependent manner. In 1998, a national survey among critical care physicians in Canada demonstrated significant variations in transfusion thresholds, highlighting the need for randomized controlled trials to determine the optimal transfusion strategy in critically ill patients [4]. This was followed by the TRICC trial, where Hebert and colleagues compared transfusion with trigger Hb of 10 to 12 g/dl (liberal strategy) with a more restrictive strategy for transfusion at a trigger Hb of 7 to 9 g/dl in critically ill patients [5]. The authors observed that while the outcome, measured as 30-day mortality, was similar in both groups (18.7 percent vs. 23.3%, p = 0.11), the restrictive transfusion strategy was associated with a significant decrease in the number of patients being transfused as well as the total number of units transfused during the study period. The results from TRICC favored restrictive transfusion strategies and elucidated potential harm with the liberal transfusion strategy. But its generalizability to various subgroups of critically ill patients remains controversial. Until further evidence becomes available, we believe we can safely set a transfusion threshold of 7 g/dl in a stable patient without comorbidities and 10 g/dl for bleeding patients and those with active myocardial ischemia. The results of the RCTs looking at different subgroups of critically ill patients are summarized in Table 12.1.

Table 12.1 Randomized trials comparing liberal versus restrictive transfusion strategies in different subgroups of critically ill patients
Full size table

Age of RBC and Transfusion Outcomes

RBCs are currently stored up to 42 days after collection as per the regulatory authority, depending upon the type of anticoagulant and additive solution used during the preparation process. During storage, RBCs undergo structural, biochemical, and metabolic changes, known as the “storage lesion.” As a result of prolonged storage, RBCs may become ineffective and accumulation of bioactive substances can also lead to unwanted biological effects. Blood transfusion services usually issue the oldest compatible RBC units available as a part of FIFO policy (first in, first out) to minimize waste of blood components; this usual practice can lead to harm in critically ill patients. Though various observational studies and systematic reviews have reported adverse outcomes associated with stored/old blood, recent RCTs have refuted the claim. As shown in Table 12.2, to date, four RCTs have evaluated the fresh vs. old RBCs or standard issue RBCs and have not shown a difference in mortality or other outcomes based on RBC age. Therefore, we believe RBC age does not have clinically relevant effects on patient condition.

Table 12.2 Randomized trials comparing fresh versus old RBC transfusion in critically ill
Full size table

Thrombocytopenia and Platelet Transfusion

Thrombocytopenia is a frequent complication encountered during critical illness, with a reported prevalence between 8.3% and 67.6% at the time of admission and up to 44.1% in patients with normal platelet counts during admission [19]. Thrombocytopenia in critically ill patients results from hemodilution, increased platelet consumption, decreased production, increased sequestration, and destruction. In addition, platelet dysfunction due to the underlying disease itself and due to medications can further add to the increased risk of bleeding in these patients. Platelet transfusions are required to treat thrombocytopenia-related bleeding and as a prophylactic measure for patients at risk of bleeding or with impaired platelet function.

In the ICU setting, 10% to 30% of patients will receive platelet transfusions, the majority of which are used as a prophylactic measure to prevent bleeding [20]. While platelet transfusions are an established trigger in thrombocytopenic patients with bleeding, prophylactic platelet transfusion in ICU setting are highly debated due to lack of evidence. Recommendations for platelet transfusion thresholds are largely based on expert opinion.

As critically ill patients are prone to bleeding and frequently undergo invasive procedures (surgery, catheters), the need for platelet transfusion should be balanced against the risks of transfusions. It is important to assess the risk of bleeding, cause and pattern of thrombocytopenia, and presence of comorbidities before making the decision to transfuse.

While hemorrhage in the presence of thrombocytopenia is an established trigger for therapeutic platelet transfusion, there is no predefined level of platelet count to be maintained in such patients. Based on expert opinion, several guidelines recommend a threshold of 50x109/ml in acutely bleeding patients. In fact, in severely injured trauma patients with massive hemorrhage, early platelet transfusion with target platelet count of 100x109/ml is recommended to prevent the coagulopathy of trauma. The degree of thrombocytopenia alone is not a prominent contributor to the hemorrhage and its consequences. Prophylactic platelet transfusions in the ICU setting are often required when there is a need for surgical or radiological intervention or if the patient is at increased risk of bleeding due to presence of additional risk factors such as fever, infection, concomitant diffuse intravascular coagulopathy (DIC), severe hepatic or renal dysfunction, and use of antiplatelet medications. Based on consensus, prophylactic platelet transfusion in non-bleeding patients are recommended at a threshold level of 10x109/l in the absence of additional risk factors for hemorrhage and 20–30x109/l for those with additional risk factors. Higher thresholds of 50x109/l for platelet transfusions are recommended if there is a possibility of platelet dysfunction, even if the patient is not thrombocytopenic. Similarly, for patients with neurological complications such as intracranial bleeding, a higher threshold of 100x109/l has been suggested. A summary of the recommended threshold for platelet transfusions is shown in Table 12.3. Though the threshold for platelet transfusion in ICU patients undergoing invasive procedures has been defined, the evidence base to guide the same is poor.

Table 12.3 Clinical indications for platelet transfusion
Full size table

Coagulopathy and Plasma Transfusion

Coagulopathy is another condition frequently encountered in the ICU, occurring in up to two-third of critically ill patients [21]. Presence of coagulopathy in critical illness can increase the risk of developing hemorrhagic complications fivefold compared to patients with a normal coagulation status [22]. To assess bleeding risk and the effectiveness of plasma transfusion, prothrombin time (PT) or international normalized ratio (INR) is most widely used [23, 24]. However, as the coagulation status of a patient is the net result of a balance between procoagulant and antifibrinolytic activity, these tests poorly represent in-vivo hemostatic potential.

FFP and cryoprecipitate transfusions are used in the treatment and prevention of hemorrhage. Approximately 13% of ICU patients will receive a plasma transfusion during their admission, 70% of which are used prophylactically prior to an invasive procedure or to correct abnormal coagulation tests [24, 25]. Despite the lack of evidence to support the use of prophylactic plasma transfusion for correction of laboratory anomalies and during low-risk procedures such as central venous catheter insertion, percutaneous tracheostomy, thoracentesis, and lumbar puncture, prophylactic FFP transfusion is still a common practice in the ICU. Table 12.4 summarizes the indications for FFP and cryoprecipitate transfusions. There is increasing concern that adverse reactions associated with such transfusions will affect the risk vs. benefit balance of prophylactic FFP transfusion.

Table 12.4 indications for FFP and cryoprecipitate transfusion
Full size table

Given the poor ability of conventional coagulation tests to predict bleeding, viscoelastic methods have gained importance in the monitoring of coagulation status, especially in bleeding patients. These tests offer numerous advantages over the conventional coagulation tests and are able to better guide transfusion in critically ill patients. These tests, e.g., thromboelastography (TEG) or rotational thromboelastometry (ROTEM), provide an overall picture of hemostasis, including coagulation and fibrinolytic pathways (Fig. 12.1). Additionally, they can be performed at the point of care giving more accurate information about the patient’s dynamic coagulation status with faster turnaround times. While the use of TEG has been shown to reduce bleeding-related morbidity and mortality in cardiac surgery and trauma patients, more clinical research is required to validate its utility in the critically ill patient population.

Fig. 12.1
A curve for T E G parameters. The highest amplitude on both sides is M A. The section on the right and left of M A are clot formation and fibrinolysis.

Specific TEG parameters represent the three phases of the cell-based model of hemostasis: initiation, amplification, and propagation

Full size image

  • R value = reaction time (s), time of latency from start of test to initial fibrin formation (amplitude of 2 mm), initiation phase, dependent on clotting factors (normal 4–8 min). Corresponding terminology for ROTEM is clotting time (CT).

  • K = kinetics (s), time taken to achieve a certain level of clot strength (amplitude of 20 mm), amplification phase, dependent on fibrinogen (normal 1–4 min). Corresponding terminology for ROTEM is clot formation time (CFT).

  • alpha (α) = angle (slope of line between R and K), measures the speed at which fibrin build up and cross-linking takes place, hence assesses the rate of clot formation, “thrombin burst” or propagation phase, dependent on fibrinogen (normal α-angle: 47–74°).

  • TMA = time to maximum amplitude(s).

  • MA = maximum amplitude (mm), represents the ultimate strength of the fibrin clot; i.e., overall stability of the clot, dependent on platelets (80%) and fibrin (20%) interacting via GPIIb/IIIa (normal 55–73 mm). Corresponding terminology for ROTEM is maximum clot firmness (MCF).

  • A30 or LY30 = amplitude at 30 mins, percentage decrease in amplitude at 30 mins post-MA, fibrinolysis phase (normal 0–8%). Corresponding terminology for ROTEM is clot lysis (CL).

  • CLT = clot lysis time (s).

Approximate normal values (kaolin-activated TEG, values differ if native blood used, and between types of assay).

Adverse Transfusion Reactions in Critical Care

Transfusion of blood and blood components is often lifesaving but can be associated with adverse effects including metabolic complications (e.g., hypothermia, acidosis), transfusion-transmitted infections (e.g., HIV, HCV, HBV), transfusion-associated circulatory overload (TACO), hemolytic transfusion reactions (HTR), febrile nonhemolytic transfusion reactions (FNHTR), allergic transfusion reactions, transfusion-related acute lung injury (TRALI), transfusion-associated graft versus host disease (TA-GVHD), nosocomial infection, and transfusion-associated immunomodulation (TRIM). TRALI, TACO, and nosocomial infections are frequently encountered in the ICU setting.

TRALI

Defined as acute non-cardiogenic pulmonary edema developing within 6 h of transfusion with a PaO2:FiO2 ratio of <300 mmHg in room air and bilateral infiltrates on a chest radiograph in the absence of left atrial hypertension. It occurs more commonly with the transfusion of cellular blood components rather than plasma-based components. Critically ill patients are susceptible with estimated incidence up to 8% in this population [26]. Mortality rates for TRALI range from 9 to 15% but can be as high as 40% in critically ill patients [27].

TACO

Defined as acute respiratory distress with pulmonary edema, tachycardia, increased blood pressure, and evidence of a positive fluid balance after a blood transfusion. Although all blood components have been implicated as potential causes of TACO, recent studies have identified FFP transfusion as a frequent cause [28]. The exact incidence of TACO is unknown but is common in critically ill patients. TACO accounted for 44% of the transfusion-related deaths reported to the UK Haemovigilance during 2010 to 2017 [29].

Nosocomial infections and transfusion

The risk of infection following RBC transfusion is related to the amount of transfused blood and RBC storage duration. An increased risk of nosocomial infection following blood transfusion in critically ill patient populations has been demonstrated in a number of studies [30, 31]. Similarly, platelet and plasma transfusions have also been associated with postoperative infection in cardiac surgery and critically ill patients recovering from sepsis. As RBCs are often administered together with plasma and platelets, it is difficult to ascertain the exact component as the causative factor.

Alternatives to Transfusion

As previously discussed, anemia in critical illness is primarily due to functional iron deficiency (presence of chronic illness) and blunted erythropoietin response. The use of alternative strategies can reduce the incidence and severity of anemia and need for RBC transfusion and may reduce morbidity and mortality. Intravenous (IV) iron was studied as an alternative treatment for anemia in critically ill patients. However, the IRONMAN study failed to demonstrate a decrease in RBC transfusion requirements, although patients who received intravenous iron had significantly higher hemoglobin concentration at hospital discharge [32]. The use of IV iron preparation does have increased theoretical risks of infections and adverse reactions.

The use of erythropoietin in critically ill patients has also been evaluated in several RCTs. Significant decrease in RBC usage was observed in earlier studies but subsequent trials failed to demonstrate a consistent effect, suggesting that the benefits of using erythropoietin became negligible once the restrictive transfusion trigger of 7.0 g/dl became the standard [33]. No differences in other patient outcomes were noted.

Tranexamic acid, an antifibrinolytic agent, has been shown in the CRASH-2 study to reduce the need for blood transfusion in the trauma setting without an increase in thromboembolic events [34]. Its use is also recommended in postpartum hemorrhage, high-risk surgery, and other non-trauma settings [35, 36]. However, utility of tranexamic in the highly variable population of critically ill patients needs further evaluation.

Case Vignette

The patient in the vignette was given component therapy transfusion based on thromboelastography. Components were titrated to the results until the bleeding was effectively controlled. Thereafter, transfusion was stopped. As a consequence of the treatment, the patient developed TRALI which was managed successfully. The indications for blood and its components in ICU are either active bleeding or prophylactic. The age of RBCs does not matter. The management of coagulopathy in critical illness should be individualized. The potential adverse effects of blood transfusion include infection, TRALI, TACO, hemolysis, graft versus host disease, and immune modulation.

Conclusion

Blood and blood products constitute major lifesaving therapy especially in critically ill patients who are actively bleeding or at risk of major bleeding. The threshold for initiation of transfusion should be based on individual factors. However, the evidence supports restrictive use in the majority of cases. The risk–benefit ratio of adverse events should be considered when making the decision to transfuse. The use of newer viscoelastic tests provides dynamic assessment and can help in rationalizing the decision for component therapy.

Take-Home Messages

  • RBC transfusion is required to improve oxygen-carrying capacity and also promote hemostasis.

  • Restrictive RBC transfusion strategy in critically ill patients is more beneficial in reducing the volume of transfusion requirement and improved patient outcomes.

  • RBC age does not have clinically significant effects on patient outcomes.

  • Platelet transfusion is indicated to treat thrombocytopenic bleeding.

  • Prophylactic platelet transfusion in critically ill should be administered after risk assessment for bleeding, cause and pattern of thrombocytopenia, and presence of underlying comorbidities.

  • FFP and cryoprecipitate transfusions are used in the treatment and prevention of coagulopathy in critically ill patients.

  • Use of point-of-care tests such as thromboelastography (TEG) or rotational thromboelastometry (ROTEM) have gained importance in the monitoring of coagulation status and can be used to guide to blood transfusion.

  • TRALI, TACO, and nosocomial infections are frequently encountered transfusion reactions in the ICU setting.

  • Transfusion alternatives such as IV iron, erythropoietin, and tranexamic acid should be considered whenever feasible.