Notice: Wiley Online Library will be unavailable on Saturday 30th July 2016 from 08:00-11:00 BST / 03:00-06:00 EST / 15:00-18:00 SGT for essential maintenance. Apologies for the inconvenience.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Blood platelets are of critical importance in the first step of normal hemostasis, a process known as primary hemostasis.1 The hemostatic system consists of a complex sequence of events designed to prevent blood loss in case of vessel wall damage. Normally, hemostasis is tightly regulated and in balance. In patients with severe liver disease and during liver transplantation, multiple changes in several hemostatic processes, including primary hemostasis, occur.1, 2 These changes are related to both prohemostatic and antihemostatic pathways disturbing the hemostatic balance and may lead to either a bleeding diathesis or thrombotic disorders.1, 3 In patients with liver failure, platelet number and function are often compromised, and during transplantation, further alterations in platelet number and function may occur.4, 5 Changes in platelet count and function may contribute to increased blood loss requiring the transfusion of platelets. However, there is no consensus on how to monitor the function of primary hemostasis in surgical patients or on the most optimal transfusion threshold. Moreover, recent studies have indicated that platelet function in patients with cirrhosis may not be as compromised as previously believed, making this issue even more complex.6, 7
Apart from the well-known role of blood platelets in blood coagulation, there is increasing evidence that platelets have various other nonhemostatic properties that play a critical role in inflammation, angiogenesis, tissue repair and regeneration, and ischemia/reperfusion (I/R) injury.8–11 All these processes may, to some extent, be involved in the (patho)physiological alterations that occur in patients undergoing liver transplantation (Fig. 1). The relevance of these nonhemostatic properties has not become evident until recently, and knowledge in this field is rapidly emerging. Experimental and clinical research points toward a dualistic role of platelets in patients undergoing liver transplantation, resulting in both beneficial and detrimental effects (Table 1). Although platelets are required to seal surgically injured blood vessels and platelet transfusion may in some cases be unavoidable, platelets have also been shown to contribute to I/R injury of the liver graft, and platelet transfusions have been associated with adverse outcome after liver transplantation.12
Table 1. The Balance of Potential Risks and Benefits of Blood Platelets in Liver Transplantation
1. Contribution to ischemia/reperfusion injury8, 9
2. Involvement in tissue repair after ischemic injury11
2. Involvement in pathogenesis of postoperative thrombosis60, 61
3. Involvement in liver regeneration (that is, small-for-size syndrome)11
3. Association of platelet transfusion with adverse outcome after liver transplantation12
In contrast with the detrimental effects of platelets, platelet activation in the transplanted liver might also have beneficial effects. Experimental studies in mice have shown that platelet-derived serotonin is involved in liver regeneration after partial resection.11 In addition, platelet-derived serotonin plays a role in hepatic repair after I/R injury.13
The aim of this article is to summarize and discuss the clinical and experimental research that has widened our understanding of the role of blood platelets in liver disease and during liver transplantation. The potential beneficial and detrimental effects of both endogenous platelets and exogenous, transfused platelets are reviewed, and directions for further research in this area are discussed.
ALTERATIONS IN PLATELETS IN PATIENTS WITH LIVER DISEASE
Primary hemostasis in patients with acute or chronic liver disease is characterized by 2 major alterations: thrombocytopenia and functional platelet defects.1, 14
Thrombocytopenia in patients with liver disease is at least partly caused by an increased sequestration of platelets in the spleen due to portal hypertension, resulting in congestive splenomegaly.15 In patients with advanced liver cirrhosis, up to 90% of the platelets can become sequestrated in the spleen. However, even in these patients, the peripheral platelet count decreases only moderately. Other mechanisms that explain the reduced platelet count in patients with liver disease are a reduced production of thrombopoietin by the diseased liver16 and a reduced platelet half-life, which is possibly related to autoantibodies.17 In patients with alcohol-induced cirrhosis, defective platelet production occurs as a result of folic acid deficiency or toxic effects of ethanol on megakaryocytopoiesis.18 Platelet consumption as a result of low-grade disseminated intravascular coagulation (DIC) may theoretically contribute to the thrombocytopenic state in patients with liver disease, although the presence of DIC in these patients is controversial.19
Functional platelet defects occur in both acute and chronic liver failure.20 Defective platelet function may be a result of an acquired storage pool defect,21 defective transmembrane signal transduction,22 decreased levels of arachidonic acid required for thromboxane A2 production in the membrane,23 decreased levels of functional platelet receptors as a result of proteolysis by plasmin,24 the presence of abnormal high-density lipoprotein,25 or a reduced hematocrit.26 In patients with cirrhosis and concomitant renal failure, a uremic thrombocytopathia may develop on top of this. Moreover, increased production of nitric oxide and prostacyclin, 2 important endothelial-derived platelet inhibitors, may contribute to defective platelet activation in vivo.27
Recent studies on the function of platelets and primary hemostasis in patients with cirrhosis have challenged the traditional concepts of defective platelet function in these patients.6, 28 Although in vitro platelet adhesion to subendothelial structures under conditions of flow is reduced, platelet deposition does not differ from normal controls when the platelet count and hematocrit are adjusted to normal levels.6 Therefore, the intrinsic platelet defects described in patients with cirrhosis appear not to be relevant under conditions of flow. In agreement with this observation, it has been demonstrated that the capacity of platelets from patients with cirrhosis to support thrombin generation is indistinguishable from that of controls, provided that platelet counts are adjusted to normal levels.28 These studies suggest that the intrinsic function of platelets in patients with cirrhosis may not be as abnormal as previously assumed.
In addition to this, compensatory mechanisms have been described for defects in prohemostatic systems and defects in platelet function in patients with liver disease.1, 7, 29, 30 One important compensatory mechanism is the substantially elevated level of von Willebrand factor (vWF) in patients with liver disease.30 Because of the high levels of vWF, the overall primary hemostatic function in patients with liver disease appears less disturbed than what would be expected on the basis of the platelet count and functional defects as observed in diagnostic tests. It has been demonstrated that vWF plasma levels can be elevated more than 10-fold in patients with cirrhosis.30 Despite subtle functional defects, these extremely high plasma levels support platelet adhesion under flow conditions better than plasma from healthy volunteers, thus compensating for reduced platelet numbers and function.30
Changes in the Platelet Count and Function During Liver Transplantation
During liver transplantation, alterations may occur in both the platelet count and platelet function, and these changes may lead to further deterioration of the hemostatic function.4, 5 Thrombocytopenia can be caused by hemodilution, immunologic reactions, or sequestration of platelets in the liver graft upon reperfusion.4, 10 After graft reperfusion, the platelet count decreases by 30% to 55% because of entrapment of platelets in the liver graft.8, 31, 32 The de novo development of idiopathic thrombocytopenic purpura after liver transplantation is a rare event but could explain a sudden decrease in the platelet count in an individual patient.33 Moreover, platelet function declines in the course of liver transplantation.8, 31, 32 Two mechanisms might contribute to deterioration of platelet function during liver transplantation. First, after reperfusion, a hyperfibrinolytic state develops because of the release of tissue-type plasminogen activator from the graft.8, 34–38 This release of tissue-type plasminogen activator, resulting in the formation of plasmin, results in proteolysis of key platelet receptors and vWF.35, 38 Second, platelets are activated after graft reperfusion (Fig. 2).10, 31, 39 In an experimental porcine liver transplantation model, it has been shown that a large number of degranulated platelets can be detected lying free in the sinusoids of the liver after reperfusion, and this suggests that platelets are partially activated.10 Furthermore, plasma levels of activation markers of platelets are significantly elevated after reperfusion.8, 10, 31 It has been suggested that mediators released by graft liver cells, damaged by the preservation period, lead to different degrees of platelet activation followed by adhesion to the vessel wall and platelet aggregation.31 Some of the initially sequestrated and partially activated platelets are released again from the liver into the circulation at a later stage after reperfusion.31, 39 These partially activated platelets may be included in peripheral platelet counts, but they will have a severely reduced hemostatic capacity. Measurement of the platelet count alone is therefore not likely to reflect the platelet-dependent hemostatic capacity.
Platelet function is not measured routinely during surgery as platelet function tests such as platelet aggregometry are time-consuming and not available as a point-of-care test. Some centers use thromboelastography to monitor perioperative hemostasis and to guide transfusion therapy, and platelet dysfunction may be derived to some extent from the thromboelastographic recording.40 However, the true value of thromboelastography is still incompletely known. Therefore, most clinicians currently rely on measurement of only the platelet count when hemostatic problems as a result of dysfunctional primary hemostasis are suspected. Although several investigators have attempted to develop laboratory methods and devices that are suitable for a rapid and reproducible measurement of platelet function, none of these devices have been able to stand the test of time or to acquire wider clinical applicability. New developments in this field are eagerly awaited.
THE ROLE OF PLATELETS IN I/R INJURY, TISSUE REPAIR, AND LIVER REGENERATION.
There is increasing evidence that platelets have physiological functions that extend well beyond their well-known role in hemostasis. Platelets have been shown to be critically involved in inflammation, angiogenesis, tissue repair and regeneration, and I/R injury8–11 (Fig. 1). All these processes may become activated to some degree in patients undergoing liver transplantation, potentially leading to pathological processes. As described previously, platelets are rapidly sequestrated in the liver graft after reperfusion. In this process, some platelets adhere to the sinusoidal endothelium, which has been activated as a result of organ procurement and subsequent cold and warm ischemia. The extent of platelet activation to the activated endothelium has been shown to correlate with organ function, both in animal models and in human transplant patients.8–10 Although this correlation could just be a reflection of the extent of organ damage and the degree of endothelial activation before reperfusion, there is substantial evidence that platelets are not innocent bystanders in this process and in fact actively contribute to the pathogenesis of organ damage and dysfunction.8, 9 The direct contribution of platelets to I/R injury was first suggested by Cywes et al.9 in experimental studies using isolated perfused rat livers. A positive correlation was seen between the duration of cold ischemia, degree of platelet adhesion to activated liver endothelium, and injury of the perfused rat liver. When livers were ex vivo perfused with activated platelets, hepatic injury was increased in comparison to reperfusion with unactivated platelets, and this indicated that platelet activation is directly responsible for injury to the liver and that the correlation between the extent of platelet deposition and organ damage is not just reflecting enhanced organ damage as a result of the cold ischemia. The interaction of platelets with activated endothelium resulting in organ damage is mediated by adhesion molecules such as selectins and integrins, which are highly expressed on activated platelets and endothelial cells. In agreement with this observation, it has been demonstrated that platelets induce apoptosis of sinusoidal endothelial cells, especially upon reperfusion.8 Experiments with isolated perfused rat livers have shown that the number of apoptotic sinusoidal endothelial cells increases by 6-fold after reperfusion of the liver with (activated) platelets.8, 41
Platelets seem to act in concert with leukocytes and Kupffer cells, and a triangular interaction between these cell types has been demonstrated in the mechanisms of reperfusion injury.41 Kupffer cells, the resident liver macrophages, interact with circulating blood cells. These cells rapidly activate after reperfusion of the ischemic liver and mediate injury in interactions with leukocytes and platelets. Platelets and leukocytes need functional Kupffer cells to mediate injury, and Kupffer cells are much less harmful in the absence of platelets and leukocytes.41
Although activation of platelets during liver transplantation contributes to I/R injury, it may also have beneficial consequences. One recently discovered example is the beneficial effect of platelet-derived serotonin on liver regeneration after resection. Serotonin or 5-hydroxytryptamine not only is known as a neurotransmitter but is also a potent mitogen that modulates the remodeling of tissue.42In vitro, it stimulates hepatocyte mitosis.43, 44 Platelets store serotonin and release it at sites of tissue injury as part of their action on hemostasis.45 About 95% of all serotonin found in blood is stored in platelets. In mice, a reduction of platelets or an impaired platelet activity results in failure of the regeneration after partial liver resection. It has been demonstrated that platelet-derived serotonin is responsible for regeneration after resection, and this indicates that platelet activation in the resected liver is required for regeneration. Although experimental data are lacking, it is conceivable that platelet-derived serotonin also plays a role in repair of the ischemically injured liver after transplantation. Moreover, platelet-derived serotonin could play a central role in liver regeneration after partial liver transplantation and in the situation of small-for-size syndrome. Platelet activation within the graft immediately after reperfusion might thus also have beneficial effects.11
RESTORATION OF THE PERIPHERAL PLATELET COUNT AFTER LIVER TRANSPLANTATION
In the early postoperative period after liver transplantation, thrombocytopenia is common.46 Potential mechanisms that contribute to postoperative thrombocytopenia are platelet consumption in hemostatic or DIC-like processes and sequestration of platelets in the liver following graft reperfusion.47 In adult-to-adult living donor liver transplantation, persistently elevated portal venous pressure and hypersplenism due to relatively small graft size have been associated with posttransplant thrombocytopenia.48 In addition, postoperative thrombocytopenia in these patients has been explained by relatively low thrombopoietin production.48 In living donor liver transplantation recipients with severe thrombocytopenia due to a small-for-size graft and persistent portal hypertension, splenectomy of splenic artery embolization has been proposed as a therapeutic option.48 In general, platelet count and coagulation factors increase steadily toward normal levels if liver function is restored in the first few days after transplantation. However, an absent rise in the platelet count is associated with increased mortality after liver transplantation.49–51 Figure 3 illustrates the postoperative platelet count in the first 15 days after surgery in 449 adult patients who underwent a first liver transplant in our center and shows a steady increase in the platelet count in patients surviving the first 90 postoperative days, whereas the platelet count remained low in patients who died within the first 90 postoperative days. A persistently low platelet count can result from increased consumption of platelets or can be caused by DIC, a disorder frequently seen in critically ill patients.52 It also can be a result of sequestration of platelets into other organs, presumably due to ongoing endothelial activation.53, 54 In general, a blunted rise in the platelet count after liver transplantation should be considered an ominous sign.41
In a small proportion of patients (up to 14.7%), the rise in the platelet count following liver transplantation is excessive, resulting in thrombocytosis.52 In 1 retrospective study, posttransplant thrombocytosis was positively associated with liver transplantation for seronegative fulminant hepatic failure, whereas there was a negative association with hepatitis C–related cirrhosis. Current evidence suggests that the thrombocytosis is transient and not associated with an increased risk of hepatic artery thrombosis (HAT).52
THE IMPACT OF PLATELET TRANSFUSIONS ON OUTCOME AFTER LIVER TRANSPLANTATION
The development of liver transplantation as a therapy for end-stage liver disease would not have been possible without the use of blood product transfusions. Apart from the obvious life-saving benefits of blood products, blood transfusions also have a negative impact on outcome after surgery.56, 57 Recently, platelet transfusions have been identified as an independent risk factor for adverse postoperative outcome.12, 58 In a large retrospective analysis in patients undergoing coronary artery bypass graft surgery, it has been shown that platelet transfusions are associated with serious adverse events.58 Patients who receive platelet transfusions during cardiac surgery are more likely to have a prolonged hospital stay, more infectious complications, and a higher rate of stroke and multiorgan failure. Using multivariate logistic regression analysis with propensity score adjustments for confounding variables, Spiess et al.58 identified a 5 times higher death rate in patients who received platelet transfusions during cardiac surgery.
The impact of platelet transfusion on outcome after liver surgery is less well described. The impact of platelet transfusion on graft survival after liver transplantation was first described by Markmann et al.59 In this study, patients were divided into 2 groups according to whether they received more or less than 20 units of platelets. Transfusion of more than 20 units of platelets was an independent predictor of poor survival of primary liver grafts.57 Nowadays, this amount of platelet transfusion has become exceedingly rare. Recently, the influence of platelet transfusion on patient and graft survival after clinical liver transplantation was evaluated in detail by de Boer et al.12 In this retrospective analysis, 26 variables, including platelet transfusions, were studied in relation to patient and graft survival in 433 adult patients undergoing a first liver transplant. The indication for transplantation, platelet transfusion, and red blood cell transfusion were identified as important risk factors predicting 1-year patient survival. Although red blood cell or platelet transfusion may be a surrogate marker for sicker patients and more complex surgery, these risk factors were independent of well-accepted indices of severity of disease, such as the Karnofsky and Model for End-Stage Liver Disease scores.12 The effect on 1-year survival was dose-related, with a hazard ratio of 1.377 per unit of platelets (P = 0.01). In a subsequent study, acute lung injury was identified as the main cause of postoperative death in patients who received intraoperative platelet transfusions (I. T. A. Pereboom et al., Platelet transfusion during Liver Transplantation is associated with increased postoperative mortality due to acute lung injury. Anesth Analg, 2008). The impact of platelet transfusion on graft survival was less pronounced. In a univariate analysis, graft survival was negatively associated with platelet transfusion, but the effect disappeared in a multivariate analysis.12 Although these studies provide strong support for a detrimental effect of platelet transfusions on outcome after liver transplantation, it is difficult to prove causality in retrospective analyses. Despite the lack of randomized studies, these findings are in agreement with previous studies performed both within and outside the field of liver transplantation, which show detrimental effects of platelet transfusions on postoperative outcome.12, 58 Although there is no good alternative therapy for platelet transfusions in the situation of excessive bleeding and low platelet count, it seems advisable to avoid prophylactic transfusion of platelets in liver transplant recipients as much as possible.
THE ROLE OF ANTIPLATELET THERAPY IN LIVER TRANSPLANTATION
In contrast to the bleeding tendency that is usually encountered intraoperatively, thrombotic complications are among the most frequent complications after transplantation.60, 61 One of the most feared thrombotic complications after transplantation is thrombosis of the hepatic artery.60 Because platelets are known to play a pivotal role in the development of arterial thrombosis, antiplatelet therapy appears to be an attractive preventive method.62 However, on the basis of the traditional concept that patients undergoing liver transplantation have a high risk for bleeding, the use of anticoagulant medication is generally avoided in these patients, and very few studies have focused on the efficacy and safety of antiplatelet drugs after liver transplantation.63–65 Although some investigators have advocated the routine use of aspirin as antiplatelet therapy after liver transplantation,61 the efficacy and safety of antiplatelet therapy in these patients have been studied in only 2 retrospective studies.64, 65 In a retrospective study, Wolf et al.65 found no benefit of prophylactic low-dose aspirin therapy in the prevention of early HAT after liver transplantation. Although aspirin was not associated with an increased risk for bleeding complications, a trend toward an increased incidence of gastrointestinal bleeding was seen in that study. Therefore, aspirin as prophylactic therapy for early HAT was not recommended by the authors. In contrast, in a more recent study by Vivarelli et al.,64 the efficacy and safety of long-term aspirin administration for the occurrence of late HAT have been described. Late HAT was seen more frequently in patients who did not receive antiplatelet prophylaxis after liver transplantation (12 out of 338) in comparison with patients with the same risk factors who did receive aspirin (1 out of 160). There were no hemorrhagic complications associated with the use of aspirin. These results suggest that antiplatelet prophylaxis may effectively reduce the incidence of late HAT after liver transplantation, particularly in patients at risk for this complication.64
Although the anticipated bleeding risk in patients undergoing liver transplantation is presumably lower than previously thought (especially in the postoperative period), we should remain extremely careful with the administration of anticoagulants or antiplatelet drugs and be aware of the fact that these drugs carry a risk of inducing bleeding complications. Prospective, controlled clinical studies in this area are eagerly awaited.
SUMMARY AND FUTURE PERSPECTIVES
Blood platelets are of crucial importance for hemostasis. During liver transplantation, a certain amount of platelets is required to maintain hemostasis; however, no evidence-based threshold values for platelet count are available. Also, no consensus exists on the appropriate use of platelet transfusion during liver transplantation. Moreover, platelet function is not routinely measured in the operating room, but as it is likely that platelet function deteriorates during liver transplantation, this information would be useful in addition to a platelet count to guide transfusion. Furthermore, it is important to distinguish between prophylactic platelet transfusion (that is, platelet count under a certain threshold, without bleeding complications) and transfusions intended to control bleeding. Reduction of the prophylactic use of platelet concentrates should be possible, as it is known that the platelet count will normally restore spontaneously after transplantation. Therefore, it might be reasonable to withhold platelet transfusions in a patient with a low platelet count but no excessive bleeding and to transfuse only if bleeding complications do occur.
Mechanisms by which platelet transfusions result in increased mortality are incompletely known. They probably include general transfusion-related problems such as infection and immunological reactions but may also be a result of the slightly preactivated state of transfused platelets, which combined with the activated state of the graft endothelium results in unwanted platelet–vessel wall interactions. Platelet concentrates are prepared by either a whole blood–derived procedure (platelet-rich plasma or buffy coat–based) or platelet apheresis.66 When platelet concentrates are prepared by apheresis, only 1 donor is used, whereas pooled material from 4 to 8 donors is used when platelet concentrates are prepared from whole blood. Although apheresis is more expensive than using whole blood, a decrease in the number of donors to which a recipient is exposed probably decreases the risk of acquiring transfusion-related complications. More detailed insights into the risks and benefits of the different production processes of platelet concentrates are needed.
Although platelets are harmful by inducing I/R damage, platelets also secrete substances such as serotonin, which may assist in the repair of ischemic damage after liver transplantation. In this respect, platelets can be considered a 2-edged sword, and a better understanding of these opposing functions as well as the role of endogenous platelets versus transfused platelets is required. Strategies to avoid platelet interaction with activated sinusoidal endothelial cells could be pursued, but this might also affect serotonin delivery within the liver. Conversely, one could envision strategies aimed at enhancement of serotonin release in the liver, such as the administration of serotonin precursors.
Platelets play a pivotal role in the imbalance between prohemostatic and antihemostatic pathways during the first days after liver transplantation, probably contributing to thrombotic complications. Because of the central role of platelets in arterial thrombosis, antiplatelet therapy, such as aspirin, seems an attractive method to avoid these complications after liver transplantation. Retrospective studies have suggested that long-term administration of aspirin after liver transplantation is safe and has a positive effect on outcome. These results open new avenues for further research. Clinical studies focusing on the safety and efficacy of antiplatelet drugs in reducing vascular thrombotic complications after transplantation are eagerly awaited.
In conclusion, platelets are both friend and foe during liver transplantation. Experimental and clinical research has widened our insight into the beneficial and detrimental effects of platelets in patients undergoing liver transplantation. More research in this area, including clinical trials focusing on the modulation of platelet function, however, is warranted.