Clinical Effects and Possible Mechanisms of Transfusion-Related Immunomodulation

Abstract

Allogeneic blood components are commonly transfused in trauma, surgery, and intensive care units and are related with adverse effects, such as postoperative infections, multi-organ failure, and mortality. The adverse effects of blood transfusions on the immune system are called as transfusion-related immunomodulation (TRIM). Many clinical trials are conducted to show the clinical effects of TRIM. They found in different clinical settings controversial results. There are many possible mechanisms of TRIM. Although until now, the exact mechanisms are not elucidated resulting in a challenge to unravel this complex interaction between immunomodulation and clinical events leading to morbidity and mortality. It has been postulated that allogeneic leukocytes are associated with the clinical adverse effects of TRIM that predominantly is observed in high-risk patients as cardiovascular surgery. Allogeneic leukocytes could activate inflammation cascade leading to adverse events in high-risk patients. Also other blood components as red cells, plasma, and platelets can play a role in the development of inflammatory complications after blood transfusions. In this review, we will discuss the clinical effects and the possible mechanisms of TRIM in relation with allogeneic leukocytes and mediators derived from allogeneic blood transfusions.

1. Introduction

Until the discovery of the ABO blood groups in the early 1900s blood transfusions were a high-risk procedure: more than 50% of the recipients of blood did not survive. The discovery of the ABO blood groups followed by the development of citrate as anticoagulant to prevent clotting of blood enabled the start of a long history of transfusion medicine. Since blood component transfusion of red blood cells (RBCs), platelet concentrates, and plasma became possible over time, blood transfusions became gradually considered as safe for the treatment of blood loss and other causes of anemia [1]. Allogeneic blood components are commonly transfused in trauma, surgery, and intensive care units (ICUs); up to 60% of these patients receives a blood transfusion [2, 3]. It is known that allogeneic blood transfusions have adverse effects which can lead to deleterious outcome in recipients. These effects of allogeneic blood transfusions are called transfusion-related immunomodulation (TRIM). The clinical effects of TRIM are most seen in high-risk patients as to cardiac surgery [4]. In this review, we will focus on the clinical effects and possible mechanisms of TRIM, especially in patients at high risk.

2. Transfusion-related immunomodulation (TRIM) in different clinical settings

The existence and possible mechanisms of TRIM are not yet fully discovered. Many factors might contribute to TRIM. Allogeneic leukocytes and (leukocyte-derived) soluble mediators in blood products were considered as most important [4]. Through filtration, the number of allogeneic leukocytes in donated blood can be reduced by more than 99.9% with a residual leukocyte count of less than 1x10⁶/L. Leukocyte-depleted blood transfusions were applied since the 1980s to reduce nonhemolytic febrile transfusion reactions, human leukocyte antigen (HLA) allo-immunization, and Cytomegalovirus (CMV) transmission in patients at risk. In early 2000s, in many Western countries, universal leukocyte depletion of allogeneic blood transfusion was implemented.

In the 1970s, it was discovered that pretransplantation allogeneic blood transfusions improved a subsequent renal allograft survival [5]. It was hypothesized that allogeneic leukocytes present in the blood components could lead to an immunosuppressive transfusion effect resulting in impaired cancer surveillance and to susceptibility for postoperative infections [4, 6, 7]. Although, two randomized controlled trials (RCTs) compared buffy-coat-poor red blood cells (RBC) with filtered RBC on cancer recurrence after colorectal surgery. Both studies found no difference in recurrence after 5 years [8, 9]. Therefore, a possible immunosuppressive effect of allogeneic blood transfusion on cancer recurrence is not found.

On the other hand, many observational studies showed an association between postoperative infections and allogeneic blood transfusions. The presumption was that the immunosuppressive effect of leukocyte-containing RBC transfusions was responsible for postoperative infections. Many RCTs were conducted in various clinical settings, evaluated postoperative after leukoreduced (LR) RBC transfusions. These studies varied as to single- or multiple-center design, clinical diagnosis, methods to document infections and proportion of transfused patients ranging between 14 and 95% and revealed different outcomes.

Several meta-analyses were performed investigating the effects of, allogeneic leukocyte-containing, buffy-coat-depleted (BCD) RBC transfusions, but these came also to controversial conclusions. Meta-analyses using intention-to-treat analyses seldom found an association between LR transfusions and postoperative infections [10]. A meta-analysis, restricted to transfused patients only, thereby excluding 36% of the study population, reported up to almost 60% reduction in postoperative infection after transfusion of LR RBC [11]. The effect of LR RBC was mainly observed in patients undergoing cardiac surgery. On the other hand, mortality was also investigated in different clinical settings. Overall, no adverse effect of leukocyte-containing transfusions on short-term mortality has been found. Only in cardiac surgery, the mortality rate was significantly decreased in patients who received LR RBC [10].

3. Cardiac surgery and allogeneic blood transfusions

Coronary artery bypass graft (CABG) surgery is a frequently performed intervention to re-vascularize the myocardium. Worldwide, approximately 1 million patients are undergoing cardiac surgery annually. The current mortality rate after cardiac surgery is low, and cardiac surgery has become a routine procedure. Although the number of patients who receive blood transfusions and the numbers of transfused blood products became lower in time, patients undergoing cardiac surgery still receive more blood transfusions compared to other surgical settings. Due to hemodilution and consumption of coagulation factors and platelets in the extracorporeal circuit, patients undergoing cardiac surgery can develop severe bleeding complications. In cardiac surgery, transfusion rate varies between 27 and 92% and estimated approximately 10% of the total blood supply worldwide [12]. Also platelets and plasma are frequently transfused to patients undergoing cardiac surgery.

After cardiac surgery, patients generally stay at an intensive care unit (ICU) for as long as mechanical ventilation and cardiac inotropic drug support is needed. Anemia is often encountered in the ICU in surgical patients and is of multifactorial origin. Besides, hemodilution by abundant intravenous infusions, decreased RBC production due to iron deficiency and inappropriate erythropoietin response due to inflammatory mediators in critically ill patients, reduced RBC survival, and increased (drug-induced) hemolysis may contribute further to postoperative anemia in ICU patients [13, 14, 15]. A prospective observational study in several ICUs found that approximately 29% of the patients reached a hemoglobin value of less than 6.2 mmol/l (10.0 gr/dl) with a transfusion rate of 37%. Of the patients with an ICU stay >7 days, 73% had received allogeneic RBC transfusions. Overall mortality was almost twice as high in patients who received RBC transfusions compared to patients who were not transfused. In critically ill patients, blood transfusions have been associated with mortality, ventilator-associated pneumonia, acute respiratory distress syndrome (ARDS), and bloodstream infections [16, 17, 18, 19, 20]. Also transfusion of platelets and plasma were reported to contribute to the development of these complications [21, 22].

The effect of allogeneic blood transfusions containing leukocytes was predominantly present in cardiac surgery. This was observed in two RCTs. One randomized controlled trial, aimed to investigate the development of HLA antibodies and postoperative infections after RBC transfusions in cardiac surgery (CABG +/− valve surgery), found surprisingly a higher mortality rate in patients receiving leukocyte-containing RBC transfusions [23]. Mortality due to multi-organ dysfunction syndrome (MODS) was the major cause of excess deaths after non-LR transfusions. In this study, patients were randomized to three different blood products; BCD RBCs were compared with two filtered RBCs: either fresh filtered RBCs before storage (FF) or stored filtered RBCs (SF). Between the two types of filtered RBCs, the mortality rate was not different. A subsequent RCT conducted in high-risk cardiac surgery (anticipating higher transfusion needs) investigated the effect of leukoreduction on the incidence of MODS but found no difference after BCD or LR RBCs [24]. However, MODS (with the presence of postoperative infections) as a cause of death occurred more often in patients who received BCD RBC [25]. Recently, in 150 patients undergoing cardiac surgery, it was found that patients receiving leukocyte-containing blood transfusions had longer ICU stay and hospital stay, and the duration of mechanical ventilation was longer. These patients had more risk for developing postoperative kidney injury [26]. A meta-analysis in patients undergoing cardiovascular surgery concluded that leukocyte-depleted blood transfusions resulted in significant reduction in postoperative infections (OR = 0.77, 95% CI = 0.66–0.91) and all-cause mortality (OR = 0.69, 95% CI = 0.53–0.90) [27]. Further, a large analysis enrolling >14.000 patients undergoing cardiac surgery and hip surgery or admitted to ICU after surgery or trauma observed after universal leukoreduction significant lower in-hospital mortality and decreased incidence of fever and antibiotic use in patients receiving blood transfusions [28]. The available findings support the standard use of LR in cardiac surgery, if universal leukoreduction is not implemented [29].

As possible explanation for these differences in cardiac surgery, it was postulated that patients undergoing cardiopulmonary bypass develop a systemic inflammatory response syndrome (SIRS), where allogeneic blood transfusions could have a complementary role. Generation of inflammatory mediators may be associated with more complex and longer surgery, whereas these patients receive also larger amounts of blood transfusions. During cardiac surgery, blood is exposed to the extracorporeal circuit, hypothermia, and ischemia/reperfusion injury. These insults are potent inducers of a stress response. SIRS usually resolves with adequate supportive therapy, and most of the patients recover. However, overwhelming SIRS can dominate and progress to MODS, which may lead to mortality [30]. Transfusion of leukocyte-containing RBCs to a patient with an already existing inflammatory cascade can contribute to an increased morbidity and mortality (second hit). It has been hypothesized that leukocyte-containing RBC transfusions can further imbalance SIRS leading to aggravation of MODS and could finally result in death [31].

4. Possible mechanisms of TRIM

Many factors present in allogeneic blood products have been proposed to induce TRIM. These can due to factors derived from leukocytes, red blood cells, platelets, and plasma [32, 33].

4.1 Leukocyte-derived factors

Clinical studies suggested that allogeneic leukocytes are responsible for the deleterious effects of (leukocyte-containing) blood transfusions. In blood products up to 5x10⁶/L, residual leukocytes can present. Probably by apoptosis, these residual leukocytes and leukocyte-derived mediators or residual leukocytes may induce immunomodulation [34]. Also after transfusion, microchimerism may be present 2 years after transfusion in 25% of trauma survivors, which can contribute to the development of immune suppression [35]. Further, interaction between residual leukocytes from allogeneic blood transfusion and recipient’s lymphocytes may result in allo-immunization [36]. Also leukocytes became apoptotic during the storage of blood products. These factors can lead to immunomodulation [37, 38]. One study in cardiac surgery showed as a marker for apoptosis an increase in Fas ligand in patients receiving leukocyte-containing blood [26]. Although a direct causal interaction between residual leukocytes and postoperative complications related with allogeneic blood transfusions remains uncertain.

Soluble leukocyte-derived factors, like cytokines and HLA molecules, can modulate recipient’s immune system. It has been suggested that pro-inflammatory and anti-inflammatory cytokines can accumulate during the storage of allogeneic blood transfusions [39]. Also degranulation products of remaining leukocytes like histamine, serotonin, elastase, and acid phosphatase can contribute to an immunomodulatory effect [40]. In a multivariate analysis, the number of contaminating leukocytes and the storage duration of RBCs were the most significant factors associated with febrile nonhemolytic transfusion reactions [41]. Further, it has been suggested that interleukin-8 may be the cause of transient posttransfusion leukocytosis in critically ill patients [42]. Few studies investigated the effect of allogeneic (leukocyte-containing) blood products on the cytokine balance. These studies compared cytokine profiles in patients receiving leukocyte-containing RBC transfusions with patients who did not receive any transfusions. In one study in cardiac surgery, an association was found between allogeneic RBC transfusions and postoperative increase of bactericidal/permeability-increasing protein (BPI), a marker of neutrophil activation, and the pro-inflammatory mediator IL-6 [43]. However, the interaction between blood products and the concentrations of inflammatory mediators in relation with the outcome of the patients is unknown. Although, in one RCT between stored and then filtered and fresh filtered (lacking soluble mediators) RBCs, no association was found in postoperative mortality. This finding suggests that there is not a causal role for soluble mediators [23].

It has been observed that allogeneic leukocytes can increase a T-helper-2 (Th-2) response and suppress Th-1 response [44]. This was supported by one RCT that measured the profiles of some inflammatory mediators investigating the differences between LD and BCD RBCs [45]. The IL-6 levels were significantly higher at arrival at ICU in patients after transfusion of >3 units BCD RBCs compared with LD. The IL-10 levels were not associated with number and type of transfusions in patients with or without complications, although higher IL-10 concentrations were associated with hospital mortality in both randomization arms. These results suggested that leukocyte-containing blood transfusions can aggravate the pro-inflammatory response after surgery.

4.2 Red blood cell-derived factors

During the storage of blood products, red blood cells alter and undergo rheologic changes, as impaired deformability, shape, and rigidity. Further, biochemical changes can occur as depletion of 2,3-diphosphoglycerate and nitric oxide scavenging, which result in impaired oxygen delivery [46, 47].

Hemolysis of red blood cells occurs during storage that can lead to accumulation of iron and heme, which can cause the formation of reactive oxygen species (ROS) and expression leading to tissue damage. Iron released during storage can lead to increased levels of non-transferrin-bound iron (NTBI) that can result in changes in leukocyte activation resulting in immunomodulation and production and release of pro-inflammatory cytokines [48]. Although, no difference in pro-inflammatory cytokine release was found in healthy individuals and prematures receiving old versus fresh red blood cell transfusions, while NTBI levels were significantly higher in patients receiving older RBCs [49, 50]. Also, phagocytosis of stored red blood cells in monocytes and macrophages can induce immunomodulation that results in production and release of pro-inflammatory cytokines [51]. Further, it was observed that microparticles from red blood cells accumulated during storage have also an inflammatory effect [52].

Several observational trials investigated the effect of storage time of red blood cells. These studies found controversial conclusions in different clinical settings [53, 54, 55]. Retrospective analysis suggested an association between transfusion of RBCs older than 14 days and postoperative infections after cardiac surgery. However, other studies could not confirm this and found also no association between mortality, infections, and hospital-stay [56]. However, others found no effect of storage of red blood cells on mortality in critically ill patients [57]. Meta-analysis concluded that the heterogeneity among available studies could not the question whether the storage of red blood cells can influence prognosis [58]. Until now, RCTs could not find any deleterious effects of transfusion of older RBCs compared with younger RBCs.

4.3 Platelet-derived factors

There are few data suggesting that platelets in stored blood products play a role in the development of TRIM. Platelets and platelet-derived particles can induce immune suppression and activation of an inflammatory response [59]. Also, platelets interact with leukocytes which results in the formation of aggregates and can involve in apoptosis and can thereby play a role in immunomodulation [60]. Leukocyte-containing RBCs contain prothrombotic soluble mediators, such as CD40L, which induce the synthesis of pro-inflammatory mediators that can further activate the coagulation system. Soluble CD40L and other cytokines released by platelets may play a key role in endothelial activation and can in turn activate the immune system leading to inflammation [61]. Whether platelets present in allogeneic red blood cell bags are involved in the development of TRIM is so far not known.

4.4 Plasma-derived factors

Plasma-derived products and all plasma-containing products can contribute to the development of transfusion-related acute lung injury (TRALI). Bioactive lipids which accumulate during the storage of plasma are involved in an immunomodulatory effect. In combination with neutrophil activation in plasma, bioactive lipids can induce TRALI [62]. TRALI is a life-threatening transfusion reaction with an estimated incidence of 1:1000 to 5000 plasma-containing blood transfusions. TRALI is defined as non-cardiogenic lung edema presenting within 6 hours after the completion of transfusion [63]. Endogenous neutrophil priming associated with the patient’s underlying illness, combined with bioactive lipids and modifiers in blood products, can result in neutrophil-induced pulmonary endothelial damage leading to capillary leakage. It has been hypothesized that the presence of leukocytes contributes to the generation of bioactive lipids during storage, enhancing the accumulation of lipid-priming agents and neutrophil-priming factors [64]. Besides leukocyte-reactive antibodies present in donor plasma, soluble factors accumulating during the storage of red cells and platelet products have been associated with TRALI [65].

5. Effect of platelet and plasma transfusions on TRIM

A substantial proportion of patients undergoing major surgery and trauma receive, besides red blood cells, also platelet transfusions. In cardiac surgery, approximately 20% of patients suffer from bleeding and receive platelet transfusion. Studies in cardiac surgery showed that platelet transfusion was associated with threefold increase in stroke and fivefold increase in mortality [66]. In another study in cardiac surgery, platelet transfusions were associated with mortality in patients with postoperative infections [67]. However, other studies found not an association between platelet transfusions and postoperative complications [68, 69, 70]. Patients who receive platelet transfusions could be more sicker and have more complex course, so transfusion of platelets could be a surrogate marker.

Plasma transfusions are predominantly transfused to patients who also receive large numbers of RBC transfusions. It has been suggested that plasma transfusions are associated with adverse outcome after cardiac surgery. A predominant role of plasma transfusions in cardiac surgery outcome was reported in one study in cardiac surgery [71]. Other studies that focused on plasma transfusions reported inconsistent findings [68, 72]. In another study, plasma transfusions were associated with mortality (with the presence of infections) [67]. This suggests that plasma transfusions (with the presence of soluble mediators) can contribute to postoperative inflammation in cardiac surgery. Because patients who need plasma transfusions often receive red blood cell and platelet transfusions, it is difficult to determine whether plasma transfusions are independent risk factors or are only confounders. Also one study found in >10,000 patients that the storage time of plasma transfusions was associated with early mortality, although late mortality was not [73].

The presented findings underscore the need for further studies to investigate the effects of all the various blood components transfused in cardiac surgery, as well as differentiate between adverse effects possibly associated with a specific blood component(s).

6. Allogeneic transfusion and thrombosis

Patients undergoing surgery or who suffer from cancer have a higher risk for the development of venous thromboembolism (VTE). Most observational studies investigating VTE ignored the role of allogeneic blood transfusion as a causal factor. Few studies found in cancer and after surgery that blood transfusions were dose-dependently associated with VTE [74, 75, 76]. One study in 1070 patients undergoing cardiac surgery found a dose-dependent association with transfusion of blood products and VTE [77]. In patients receiving red blood cells, the incidence of VTE was 16.7% and patients who received also plasma transfusions, this was almost doubled. The hypercoagulability and immobility after surgery could result in the activation of coagulation system, and in combination with blood transfusions this could result in higher risk in VTE.

It has been shown that platelets contain prothrombotic soluble mediators, which interact with leukocytes preceding apoptosis subsequently producing microparticles with procoagulant activity [60]. Leukocyte-containing RBCs contain soluble mediators, such as CD40L, which induce the synthesis of pro-inflammatory mediators that can further activate the coagulation system [58]. Recently, one study found in the bronchoalveolar lavage fluid, besides an increase in pro-inflammatory mediators IL-8 and TNF-alpha and also an increase in trombine-antithrombin complex (TATc), indicating the activation of the coagulation system in the lung [78]. Another study found an increase in levels of von-Willebrand-factor antigen in critically ill patients receiving a RBC transfusion. The von Willebrand factor antigen has a causal role in the activation of endothelium and also in the development of thrombi formation [79]. After cardiovascular surgery, the levels of plasma tissue factor are persistent high for 30 days, which can increase the risk of VTE [80]. In combination with plasma-derived products in stored blood products, this can result in an increased activation of the coagulation system.

7. Interaction between inflammation and coagulation associated with blood transfusions

Patients undergoing major surgery or after trauma have a higher risk for the development of both inflammation and thrombo-embolic complications. In these patients, both the inflammatory response and the release of pro-inflammatory mediators lead to an activation of the coagulation system and downregulate the anticoagulant system [81]. Activation of the coagulation factors can in turn activate inflammation. This may enhance the development of infections and microvascular thrombi [82]. Both thrombi and infections can be involved in the development and aggravation of MODS [83]. Further, by increasing the circulating red blood cell mass and vascular rheologic deformations by red blood cell transfusions, this process can be more pronounced. Also plasma and platelet transfusions can aggravate the activation of coagulation and inflammatory response.

There is a possible association between allogeneic blood transfusions and the formation of thrombosis, as a factor aggravating VTE and having a role in the development of MODS which can lead to increased mortality in patients at high risk as in cardiac surgery (Figure 1). This complex interaction and the role of allogeneic blood transfusions should be investigated in detail.

Figure 1.

Association between blood transfusions inflammation, coagulation and MODS.

8. Conclusions

Nowadays, allogeneic blood transfusions are unavoidable. Although blood transfusions are safe today, there are still concerns about transfusion-related complications. The immune-mediated complications of blood transfusions are referred as transfusion-related immunomodulation. The effects of transfusion-related immunomodulation are mainly observed in major cardiovascular surgery. Patients undergoing cardiovascular surgery are transfused with large numbers of red cells, plasma, and platelets. Although, the exact mechanisms of these immune-mediated complications are still not elucidated. The past studies showed that allogeneic leukocytes in blood transfusions play a pivotal role in the development of transfusion-related immunomodulation. Nowadays, in many countries, leukocytes are removed from transfused blood components, although the discussion about the immunologic effects of blood transfusions is continuing. In addition, by the activation of inflammatory system and release of mediators, the storage of blood components and transfusion of plasma and platelets can contribute to the development of transfusion-related immunomodulation. Further, it has been suggested that allogeneic blood transfusions are associated with the development of thrombo-embolic events in high-risk patients as cancer and surgery. It has been suggested that allogeneic blood transfusions can activate an inflammatory response and a coagulation response. Both cascades can result in worse outcome in a high-risk patient. Probably, the term transfusion-related immunomodulation is not sufficient for both inflammation and coagulation activation after blood transfusion. The term transfusion-related inflammation and coagulation (TRIC) could be more comprehensive and includes the activation of both cascades. More research (clinical and laboratory) are needed to unravel the effects of allogeneic blood transfusions in the activation of the inflammatory and coagulation cascades.

Conflict of interest

The author declares no conflict of interest.