Coagulopathy: Pathophy: Evaluation, and Treatment Bubu A. Banini and Arun J. Sanyal Keywords Bleeding - Rotational thrombelastogram ‘Thrombclastography Clotting Hypercoaguability - Thrombosis Abbreviations aT Antithrombin CVP Central venous pressure DDAVP Desmopressin EACA —_Epsilon-aminocaproic acid EPCR Endothelial protein C receptor FFP Fresh frozen plasma INR International normalized ratio PAI-1 Plasminogen activator inhibitor Pr Prothrombin time PIT Partial thromboplastin time RBC Red blood cell TACO Transfusion associated circulatory overload TAFL Thrombin activatable fibrinolysis factor TEG —Thromboelastography TF ‘Tissue factor B.A. Banini, MD, PRD ‘A.J. Sanyal, MD, MBBS (=) Division of Gastroenterology and Hepatology, ‘Department of Internal Mecicine, Virginia ‘Commonwealth University, Richmond, VA, USA ‘e-mail: babu banini@veuhealth org; arunsanyall® vouhealth org (© Springer Internation TFPI Tissue factor pathway inhibitor TM ——Thrombomodulin {PA Tissue plasminogen activator TRALI Transfusion related acute Tung injury TxA Tranexamic acid VWF von Willebrand factor Introduction Hemostasis consists of processes that promote coagulation and those that favor fibrinolysis. Both processes are essential to creating clot while focal izing thrombosis to the site of injury and prevent- ing uncontrolled thrombotic extension, In liver disease, changes occur in both anti- and pro-hemo- statie mechanisms, leading to a rebalanced coagu- lation system. In cirthosis, the rebalanced state can be disrupted by several factors including stasis, portal hypertension, dysfibrinogenemia, produc tion of endogenous heparinoids, platelet and endo- thelial dysfunction, renal failure and infection, Whether planning an invasive procedure, major surgery or transplantation, there is much dilemma in how to properly teat these patients and their ccoagulopathie status, Tis chapter will explore the balance of hemostatic pathways, and review the defects that occur in progressive liver disease. We vill also discuss how to evaluate and treat coagu- lopathy in this patient population, with specific attention given to application towards liver transplantation Publishing AG, part of Springer Nature 2018 ay G. Wagener (ed), Liver Anesthesiology and Crtcal Care Medicine, Inups:fdiorg/10-1007978-3-319-64298-7_151 B.A, Banini ang A.J. Sanyal Primary Hemostasis ‘The initial step in the hemostatic pathway occurs by formation of the platelet plug. Platelet aggrega- tion creates the scaffolding on which thrombosis ‘can then occur. When the vessel wall is damaged, subendothelial collagen is exposed to von Willebrand factor (VWF) in the serum. VWF binds to the site of injury, momentarily interacting with platelets expressing glycoprotein GPIb (Fig. 15.1a). This slows the flow of platelets until a ‘more lasting attachment is made between the ‘exposed collagen and platelet-expressed receptor 2B and glycoprotein VI, or platelet integrin UTP and fibronectin with collagen (Big. 15.1b). Glycoprotein VI on the platelet surface initiates aN GDN \@> ZL a La o transmembrane signal, allowing activation and release of ADP, thromboxane A2, and alpha and dense granules by the platelet (Fig. 15.1c). Platele-platelet interaction via integrin llbp3 can then occur, leading to further platelet aciva- tion and ageregation. In the meantime, the coagu- lation cascade initiates, leading to platelet stabilization. Secondary Hemostasis Many interactions occur simultaneously at the site of endothelial damage. As platelets aggre gate, the coagulation cascade initiates at the platelet surface, forming a fibrin clot and rein- inn DOBLE srraan Ss mt © “ ruiitaioson “ misses Fig. 15.1 (@) Subendothelisl collagen binds to von Willebrand factor (VWF), which momentarily interacts with platelets expressing glycoprotein GPIb.(b) This pro- cess slows the flow of platelets to create a more lasing attachment between the collagens and platelet-expressed receptor 42P1 and glycoprotein VI of platelet integrin Pralltiercion va sls lTbp3 and floronectin with lagen. (6) Glycoprotein VI fn the platelet surface initiates a transmembrane signal, allowing for the release of ADP, thromboxane A2, and alpha and dense granules by the platelet Platelet agerega tion can then occur15 Coagulopathy: Pathophysiology, Evaluation, and Treatment Ws forcing platelet aggregation. When the endothe- lium is damaged, tissue factor (TF) is released into the bloodstream, binding to Factor VI, initi- ating the thrombin burst. This initial generation of thrombin promotes maximal platelet activa- tion [1], as well a activation of additional coagu- lation cofactors (Fig. 15.2). While this is not ‘enough fo generate a fibrin clot on its own, it primes the clotting system for a burst of platelet aggregation by activating Factors V, VIII, and XI Fig. 15.2. Endothelial damage—aissue factor (TF interaction with Factor Vi initiating the thrombin bust, promoting platelet sctvation, and activation of addtional ‘coagulation cofactors Secor Intrluminal blood flow con the platelet surface [2]. Factor XI activates Factors IXa and VIIIA, which then come together to form the FIXa/FVIILA, tenase complex. ‘The tenase complex cleaves factor X into an activated form (Xa). Factor V is activated by FXa, This creates the fzst sufficient amount of thrombin (Ila) to generate fibrin and stabilize the platelet plug. This is known as the “propa- gation” phase of thrombin generation (igs. 15.3 and 15.4) FIL converted into Fla ndary Hemostasis: The Iniation Phase —___ Fr —_—_—____, vu Prothrombin (I) x x < ca ‘Thrombin lla) ka vila Seconds fry Hemostasis: The Ampitication Phase Intraluminalbiood flow cation Fig. 15.3. Ampli phase—the tenase complex, FIXSEVIIA, ‘leaves factor X into an activated form (Xa). FXa activates Factor V, which eats the frst ulfcient amount of xis vl Vv thrombin (la) to generate fibrin and Wr Stabilize the platelet lug collagen178 B.A, Baniniand A.J Sanyal (thrombin), in tun cleaving fibrinogen into fibrin, which forms a strong meshwork to pro- ‘mote clot stability and thrombosis. ‘The coagu- lation cascade only emphasizesthe procoagulant factors of the hemostasis. Equally important to understand are those steps which provide bal- ance and inhibit the prothrombotic steps of the coagulation cascade. Fig. 154 Propagation phase—the activity of the congulation cascade Inhibition and Fibrinolysis in Coagulation Hemostasis is composed of “forward” driving forces that promote coagulation, and those that reverse” the process to favor fibrinolysis. Both forces maintain a balance in order to localize thrombosis to the site of injury and prevent uncontrolled thrombotic extension. Thrombin (factor la) generation is directly inhibited by ts sue factor pathway inhibitor (TFPI) and ant: thrombin (AT) (Fig. 15.5). TFPI inactivates ‘Secondary Hemostasis: The Propagation Phase continues to generate thrombin, creating a stable thrombin clot Intraluminal bloed flow efdatheiom ——— > => <=> eadoT Ts ood ollagen AT ATFP! Limit Thrombosis to Endotneial Surface ig. 15.5 Tissue factor pathway inhibitor (TFPD inactivates factors Vil and Xa, inactivating factor Ta (¢hrombin) generation. ‘Also, antithrombin inativates dombin, ‘TFPFis active ony in serum, unable to inhibit, athe cellular surface, ‘Tas localizes thrombin ‘generation to the surfaces of platelets and cendotheliam where ‘damage is present ‘eneathelum > - Aoitvombin —ondeviatu ‘ollagen15 Coagulopathy: Pathophysiology, Evaluation, and Treatment Ww factors Vila and Xa, and AT inactivates throm- bin. However, TFPI is active only in serum, unable to inhibit at the cellular surface. This localizes thrombin generation to the surfaces of platelets and endothelium where damage is pres- cent (5, 6) ‘The vitamin K-dependent factors Protein C and S further regulate the coagulation cascade, Protein C is a protease [7] whose activity is ‘enhanced by protein S, and together, they inhibit factors Va and VIIIa, Protein C is localized to the ‘endothelial cell surface by the endothelial protein C receptor (EPCR) [8], further localizing throm- bin generation to the site of damage. If thrombin escapes the site of injury onto intact endothelial cells, it will be bound to the endothelial surface receptor thrombomudulin (TM), forming a thrombin/TM complex. This complex can no lon- ger carry out normal coagulant functions [9] and activates protein C to bind protein S, inhibiting clot formation (Fig. 15.6). Other mechanisms further reverse fibrin production through fibrinolysis. ‘These latter mechanisms (Fig. 15.7) utilize factors which include tissue plasminogen acti- vator (HPA), plasminogen activator inhibitor Protein C and S: Anthrombotic Activity e Na trombomedulin i] Protein © receptor endothelium Fig. 15.6 Protein Cis protease, enhanced by protein S, ‘which together inhibit factors Ve and VIIa, Protein C is localized to the endothlil cel surface bythe endothelial protein C receptor (EPCR), Thrombin escaping the site of Fibrinolysis: Clot Reversal Thrombin xi ~o. Fin Degradation Products (FDP) Fig. 15.7 FibrinolysiswPA released from endothelial ‘els, macrophages, and renal epithelial cells, activates plasminogen to plasmin. FXII inhibits the degradation of fibrin into fibrin degradation products. Thrombin actvat- Fibrinogen ~~» Fibrin {injury is bound tothe endothelial surface receptor throm bbomedulin (EMD, forming a thrombin/TM complex, and losing te ability to eary oot normal coagulant functions ©, Pan 9, rea “4 TAF uPA 8 “ <—— Plasmin <— Plasminogen t Q Piasmin inhibitor able Mbrinolysis factor (TAD inactivates the conversion of plasminogen into plasmin, by cleaving the C-terminal lysine and srpinine residues om {PA and plasminogen pre- venting their binding to one anotherv8 B.A, Baniniand A.J Sanyal (PAI-1), plasminogen, alpha2-antiplasmin, his- tidine-rich glycoprotein, and factor XIII, all of which except tPA and PAI-1, are synthesized by the liver [10]. tPA is released from endothelial cells, macrophages, and renal epithelial cells, and activates plasminogen to plasmin, Plasmin is an enzyme capable of degrading fibrin into soluble fibrin degradation products. This pro- ccess is regulated by inhibitory factors, including FXIL, which inhibit the degradation of fibrin and stabilize the fibrin meshwork, Thrombin activatable fibrinolysis factor (TAF inactivates the conversion of plasminogen into plasmin by cleaving the C-terminal lysine and arginine resi- dues on tPA and plasminogen and preventing their binding to one another. ‘When evaluating the effects of liver disease on the coagulation cascade, these inhibitory factors are often not considered. Unfortunately there is 1no simple way to test for their activity at the bed- side. They are a major contributor to the overall ‘coagulation status of the liver disease patients and need to be considered before treating coagulopathy. Hemostasis in Liver Disease After reviewing normal hemostasis, we can explore how advanced liver disease affects this process Because hepatic parenchymal cells synthesize many of the pro- and anticoagulant proteins involved in coagulation, it is easy to understand how liver disease disrupts hemostasis, Recent stud- ies have led toa greater understanding ofthe ‘rebal- ance’ of the coagulation system in liver disease, involving increased levels of circulating von Willebrand factor and factor VIII, and reduced syn- thesis of anticoagulant factors (11, 12]. However, in cimhosis, the rebalance is unstable and can be disrupted by several factors including stasis and portal hypertension, dysfibrinogenemia, produc- tion of endogenous heparinoids, platelet and endo- thelial dysfunction, renal failure, and increased susceptibility to infection Coagulation Cascade and Liver Disease The coagulation cascade is comprised of redun- dant steps that keep the entire system in balance. Ina healthy individual, only 20-50% of normal levels of procoagulants are needed to achieve hemostasis [13]. Thus, there is a fair amount of overlap between pro- and anticoagulant factors that provide a buffering system for hemostasis in healthy individuals. That buffering system is ten- uous in the liver disease patient and becomes increasingly difficult to balance as small changes in factor levels can lead to significant changes in the entire system, PT and INR have been heavily relied upon to assess the degree of coagulopathy in liver patients, This reflects a shortcoming in ‘our understanding of hemostasis and barriers in ‘our testing strategies. As PT and INR are & met measure of “procoagulant” factors, they dist {gard the effect of liver disease on “anticoagulant” factors. In liver disease, all procoagulant factors except FVIII are reduced. In addition, the antico- agulant factors such as protein C and $ are also reduced [14, 15]. The reduction of anticoagulant factors seen in liver disease is not reflected in PT ‘and INR measurements, Thus, in this rebalanced state of hemostasis, thrombin generation may actually be normal in the setting of increased PT, PTT, and INR [16], emphasizing their inade- quacy in evaluation coagulation status in liver disease. In circhosis, FVIII and other procoagu- lants including vWE can be increased [17] PTrand INR only assess one aspect of the coag- ulation cascade—namely vitamin K dependent factors FI, VIL, IX, and X—as these tests were originally developed to measure the therapeutic effects of drugs like warfarin that affects vitamin K dependent factors. In vivo, hemostasis in liver disease involves reduced levels of fibrinogen, pro- thrombin, the vitamin K dependent factors, as well as protein C and S and other anticoagulants. Moreover, factors including changes in platelet function, fibrinolysis and endothelial function all, impact hemostasis in liver disease patients15 Coagulopathy: Pathophysiology, Evaluation, and Treatment 9 Platelet Function in Liver Disease In primary hemostasis, the initial platelet plug provides the scaffolding for the coagulation cas- cade and thrombin generation. Impaired primary hemostasis has traditionally been linked to abnor- malities of platelet number and function in liver disease [15]. Platelet numbers are decreased due tothe effects of portal hypertension and increased sequestration in the spleen, and with worsening liver failure, thrombopoietin levels decrease [18, 19]. Hepatitis C, alcohol toxicity, and nutritional {folic acid deficiency compound this problem by depressing megakaryocytopoiesis (20-22. The role disseminated intravascular coagulation (DIC) as a cause of thrombocytopenia in these patients is contentious [23]; however, low-level consumption associated with DIC_ possibly decreases platelet life span as wel [24] Reduced platelet function further complicates impaired primary hemostasis, and there is strong ‘evidence that platelet aggregation is reduced [25- 27], Platelet activation is affected by both intrin- sic and extrinsic stresses, Intrinsically, decreased thromboxane A2 synthesis, altered transmem- ‘brane signaling and a eduction in glycoprotein Th and platelet integrin plTbB3 reduce platelet activa- tion [27-32]. Extrinsically, elevated levels. of nitrie oxide and prostaeyelin inhibit platelet fune- tion [33] a6 a result of endothelial dysfunction “The platelet phospholipid membrane may also be affected by abnormal high density lipoprotein particles in plasma [34]. Moreover, in liver dis- ‘ease, blood flow defects occur, and these may be compounded by reduced hemoglobin [35] Iris still not clear how these abnormalities contribute to bleeding time as elevated levels of YWE (and decreased levels of ADAMTS-13, a metalloproteinase that cleaves VWE) may com- ppensate for the impairment [12]. Infact, the ele- vation of VWF may lead to a higher rate of thrombin generation and clot formation [12). It appears that in liver disease, thrombin generation is not necessarily negatively affected and that under physiologic conditions of flow, platelets from a patient with cirrhosis can interact with collagen and fibrinogen as long as platelet count and hematocrit are adjusted to levels found in healthy patients [36, 37] Hyperfibrinolysis in Liver Disease ‘The role of hypertibrinolysis in patients with cit- rhosis is controversial and widely debated in the current literature [38]. Whether hyper- or hypofi- brinolysis is occurring, itis agreed that it compli cates the picture [3942]. Hypertibrinolysis seems to be more problematic as liver disease progresses [43-46], and low-grade fibrinolysis has been shown to occur in 30-46% of patients with end-stage disease [47]. Plasminogen, alpha2-antiplasmin, histidine rich-glycoprotein, factor XII, and TAFI [48-57] are all produced in the liver, and therefore their levels are reduced in liver disease. Howev‘ increased levels of tissue plasminogen activator (PA) and plasminogen activator inhibitor-1 (PAI- 1) are present as these are not synthesized in the liver (10). tPA is elevated most likely due to reduced hepatic clearance [58, 59]; PAIL levels seem to be correlated with the clinical stage of the liver disease, PAI-I is elevated in patients with chronic, smoldering liver disease [4 60], and decreased in patients with severe liver failure (44, 61, Patients with acute liver failure have a higher circulating amount of acute phase reactant PALI, and more ofa shift towards hypofibrinoly- sis [41], Therefore, theoretically, increased fibri- nolysis should occur in patients with severe liver failure who have an increased pool of tPA, and depressed levels of PAI-1 and alpha2-antiplasmin to balance it Decreased levels of TAFT have also been linked to hyperlibrinolysis in cirshosis (62). Colfuct and colleagues [63] demonstrated that TAFla genera- tion was low in cirthosis due to decreased levels of TAFI, suggesting that depleted TAFla was a sig- nificant contributor to hypertibrinolysis. However, Lisman et al. (64] came to the conclusion that TAFT deficiency was not a significant contributor to hyperfibrinolysis in cirrhosis. This disparity180 B.A, Baniniand A.J Sanyal may be due to the inability to test and measure global fibrinolysis, and newer methods of testing ‘global fibrinolysis confirmed the presence of hhyperfbrinolysis in chronic liver disease [65]. Fibrinolysisis important not necessarily because ofits potential to initiate bleeding ina liver disease patient but because of the important role it plays in delaying primary and secondary hemostasis, con- ‘ributing to the severity or recurrence of bleeding ‘events. In liver transplantation, many studies report ‘enhanced fibrinolytic activity during the anhepatic stage [66]. The lack of tPA clearance and the reduc tion of alpha2-antiplasmin may be responsible for this enhanced fibrinolysis (67]. After liver trans- plantation, fibrinolysis may persist for a prolonged time especially in case of early allograft dysfunc tion [67~69]. An inital rise in tPA during the anhe~ patic stage is followed by further increases after reperfusion in 75% of patients [68, 70]. The recog- nition, monitoring and treatment of fibrinolysis in moderate to severe liver disease are important and will be addressed later in this chapter. Endothelial Dysfunction and Liver Disease Sinusoidal endothelial cells produce and release vasoactive substances that regulate intrahepatic vascular resistance [71]. Endothelial dysfunction is thought to be due to a defective vasodilatory response to acetylcholine and insufficient endothe- lial NO synthase to produce NO [72-74]. Increased. production of Thromboxane A2 also leads 0 increased intrahepatic resistance in advanced liver disease (75, 76]. Portal hypertension is the liver’s response to its inability to accommodate fluctua- tions in inereased portal circulation, As portal hypertension develops, deleterious ‘processes such as splanchnic vasodilatation occur ‘The endothelium in this system responds by pro- ducing more NO resulting in arterial dilatation and a hyperdynamic circulation which is related to complications such as variceal bleeding, asci tes, hepatorenal syndrome, and hepatopulmonary syndrome [77-79]. There are several proposed. ‘ways to monitor the response of the endothelium in liver disease, and these will be addressed later. Evaluation of Coagulation Bleeding Time Bleeding time is performed by inflicting a stan- dardized cut on the volar aspect of the forearm ‘while applying a blood pressure cuf to the upper arm, There are several limitations to the repro: ducibility to the test including the skill of the technician, skin thickness, ambient temperature, and endothelial dysfunction [80-82]. Bleeding time is not well validated [81], and studies using desmopressin as treatment showed improved bleeding time but no effect on risk of variceal bleeding in liver disease patients [82-84]. The lack of correlation between bleeding time and risk of bleeding in liver disease patients makes the test not very helpful in assessing coagulation, Platelet Function Analyzer: 100 {PFA-100) The platelet function analyzer-100 (PFA-100) is an vitro test that provides a quick way to quanti tatively evaluate primary hemostasis under shear stress, The test measures platelet adhesion as blood flows through a collagen membrane under the draw of a vacuum, The time it takes to occlude the channel in the collagen membrane isthe mea- surement of platelet adhesion [80, 85]. One study demonstrated that closure time was decreased if hematocrit was normalized in the blood of liver disease patients [26] however in general the PF? 100 has not been considered overwhelmingly helpful and is rarely used to platelet function. Prothrombin Time ‘The prothrombin time evaluates the extrinsic path- ‘way of the coagulation cascade and is responsive to deficiencies in factors X, VIL, V, I and fibrino- gen. The test was developed by Armand Quick ‘and measures the time it takes a blood sample to clot once thromboplastin and calcium chloride are added [86]. Results are measured in seconds and15 Coagulopathy: Pathophysiology, Evaluation, and Treatment 181 ‘commonly standardized using the International Normalized Ratio (INR). ‘The INR was developed. as a way to account for differences in reagents (hromboplastin) across different laboratories, and. was originally used to standardize treatment of patients using vitamin K antagonists like warfarin, The use of INR is an imperfect system and in cou- madinized patients, a variability of 13% has been ‘observed depending on where lab samples are obtained [87]. This variability is accentuated in liver disease patients and is accentuated in liver disease patients and mean INR variations increase further with advanced liver disease [88-91] Not only are PT and INR variable but they also reflect an incomplete picture of coagulopa- thy in liver disease and are unreliable in this patient population [92-96]. This test for example does not reflect the parallel depletion of protein C and S in vivo [16], and is without sufficient levels of thrombomodulin for thrombin-mediated acti- vation of protein C [97]. PT is also limited by its inability assess the role of platelet and endothe- lial dysfunction, Activated Partial Thromboplastin Time ‘The activated partial thromboplastin time (aPTT) assesses the intrinsic pathway of the coagulation ‘cascade and is increased with deficiencies of all ‘coagulation factors except for factors VI and XIII, It represents the time (in seconds) for phos- pholipids, representing the platelet membrane, 10 ‘generate a thrombus by activating factors like factor XII. It is often used clinically to monitor the anticoagulant effects of heparin in patients, though there is no standardization between labo- ratories. As previously mentioned, relying solely ‘on this test to evaluate coagulation in liver dis- ‘ease patients is fraught with difficulties ‘Thrombin Generation Test This test utilizes tissue factor and phospholipids to trigger thrombin generation and is arguably the closest representation of what occurs in vivo. ‘Thrombin generation test Timeto peak peak Inactivation phase thrombin me Fig. 15.8 Thrombin generation test—tssue factor (TP) and phospholipids are added to plasma to generate throm binand wigger coagulation, AUC area under curve) =ETP Endogenous thrombin potential), or the amount of work ‘hat can potentially be done by thrombin. Proposed poten tial of this testis to quantitate how much and how long thrombin i ative Thrombin, a potent platelet activator and phos- pholipids, representing the platelet surface, feed forward to allow for explosive thrombin genera- ‘ion. In the thrombin generation test (Fig, 15.8), tissue factor and phospholipids are added to trig ‘ger coagulation, and thrombin generation is plot- ted over time, The initial part of the curve is the lag time, while the time from tissue factor addi tion to peak of thrombin is referred to as time 10 peak, The area under the curve (AUC) measures. the effectiveness of thrombin generation in a sys= tem where both pro- and anti-coagulants are examined as they operate in plasma, Reliability of the thrombin generation testis yet to be deter ‘mined but may have acceptable levels of varia tion, Further clinical studies are needed to evaluate the usefulness of the thrombin genera- tion test as it applies to liver disease and transplantation Thromboelastography ‘Thromboelastography (TEG) provides a graph- ical representation of the viscoelastic changes that occur during coagulation in vitro (Fig. 15.9). A stationary pin is introduced into a sample of whole blood that oscillates back and.182 B.A, Baniniand A.J Sanyal Fig. 15.9 ‘Tiromboclastography (EGR reticts coagulation factor and platelet activites: K reflects activity of Aibrinogen, Factor I, and effects of hematocrit; erepresens 1.5 and those with an INR 1.S who did not receive plasma subsequently did not incur any more RBC transfusion when compared to patients with an INR 50 x 10*/L without evidence of bleeding should also not be trans15 Coagulopathy: Pathophysiology, Evaluation, and Treatment 85, fused perioperatively, as these platelet levels are not correlated with bleeding risk in invasive pro- cedures such as paracentesis [136, 137]. Dosing ‘of platelets should be tailored for each individual patient and guided by bleeding, TEG, and whether ‘or not a consumptive platelet process is present. Endogenously produced platelets typically sur- Vive around 10 days; however, transfused platelets do not last this long, with further reduced life spans if platelet consumption is present. Patients ‘with platelet levels below 100% 10° have shorter platelet life spans when compared to patients with levels greater than this 138} A ssingle dose of random-donor platelets contains approximately 3 x 10! platelet, sus- pended in 50-70 mL. of plasma. The optimal dose specific to liver patients has not been ‘determined and often has more to do with indi- vidual patient requirements and availability, as platelets are limited resource (139-143). The standard dose prescription for platelets is 10 mL/kg, with a maximum dose of five ran- ‘dom donor platelet units. Single donor platelet apharesis units contain at least 300 x 10® plate- lets, suspended in 200-400 mL of plasma [144] ‘The hematology literature has found litle dif- ference in eflicacy between different platelet preparations prepared from random donors ver- sus those collected by apharesis (145, 146}. However, platelets obtained by apharesis are ‘obtained from one donor and decrease the ine dence of immune-mediated refractoriness. ‘They are used preferentially in patients who may be refractory to random donor platelet units, Platelet transfusion is recommended When patient platelet levels drop below 20-50 x 10°A. oF if platelet count cannot be ‘obtained and there is evidence of microvascular bleeding and coagulopathy unless viscoelastic testing can be obtained Procoagulant Drugs and Liver Transplantation A limited amount of information is available on pharmacologic alternatives to blood product use, Agents in this class target hemostasis, coagula- tion, and fibrinolysis, and include DDAVP, apro- tinin, lysine analogues, and recombinant activated factor VII (#FVIa). Desmopressin DDAVP or |-deamino-8-D-arginine is a synthetic analogue to vasopressin. It acts by releasing fac tor VIl and VWE from endothelial storage pools into the serum, The effect is rapid and short-lived and has a decreased response over repeated use of the drug without adequate time for storage pools to replenish. This agent has been used suc- cessfully in the treatment of minor to moderate bleeding during surgical procedures for patients with von Willebrand deficiency or hemophilia A. Dosing is usually 03 ugkg intravenously infused over 20 min, ‘There is very litte literature supporting its use during a major operation such as liver transplan- tation, however, intranasal DDAVP was shown to be equally effective as blood product transfusion in achieving hemostasis in cirthotic patients undergoing tooth extraction [147]. In this study, patients were matched evenly and there was no difference between groups transfused with plate- lets versus those given DDAVP, including MELD score, number of tooth extractions, and coagula- tion profile. For minor to intermediate procedures like tooth extraction, patients receiving DDAVP received no rescue transfusions after the proce- dure, whereas one platele-transfused patient required rescue transfusion [147]. Outcomes ‘were considered the same between the 1wo groups. Results with DDAVP have been favorable in cardiac surgery and for patients on cardiopul- monary bypass [148, 149]. Patients undergoing complex spinal fusion however did not show reduced blood Joss when treated with DDAVP versus placebo [150]. Moreover, DDAVP infu- sion did not improve hemostasis in cirrhotic patients [151]. While DDAVP should provide some theoretical benelit to liver disease patients undergoing transplant surgery, we currently cannot recommend routine use of DDAVP as a hemostatic agent.186 B.A, Baniniand A.J Sanyal Lysine Analogues Epsilon-aminocaproic acid (ACA) and. ‘ranexamic acid (TxA) are synthetic analogues of the amino acid lysine, which competitively block lysine-binding sites on plasminogen to limit bhyperfibrinolysis. The Cochrane group meta analysis found aprotinin and lysine analogues 10 have a nearly similar effect on limiting blood transfusion [152], and a recent randomized con. trolled trial of liver transplant patients failed 10 demonstrate a difference between TxA and apro- tinin with regards to transfusion requirements [134]. BACA was first shown by Kang et al. 10 decrease the amount of residual bleeding during the hyperfibrinolytic state after liver transplanta. tion in 20 of 97 liver transplant patients [153]. A. standard dose of | gram of EACA bolus was used. ‘A more recent study compared EACA, TXA, and placebo used as prophylaxis to decrease transfu- sions during and after transplant and found that TXA was beneficial over EACA. and placebo, with no difference in RBC transfusion sparing effect between EACA and placebo [154]. Dosage for TXA has varied greatly, and different studies have shown reduction in transfusion at dosages of 10 and 40 mgfkg-'/h~! 154, 155} Limited research of both of these agents makes it difficult o recommend specific dosage, and further trials should be conducted. Also, ‘comparative data is extremely limited between these two agents, and itis difficult to recommend using one over the other. Recombinant Factor Vila Recombinant Factor Vila (@FVHla) has been approved for use in hemophilia patients with inhibitors in both surgical and non-surgical set- tings. The agent is very similar to endogenous FVIla, and is thought to enhance thrombin gea- ration at sites of endothelial injury. Two mecha: nisms have been suggested in FVM driven thrombin generation: a high concentration to tis sue factor (TF) and rFVIla accumulate at the site of vascular damage [156], which in turn activates factor Xa and triggers thrombin generation (157, 158]. The second proposed mechanism involves EVI binding directly to the platelet surface and. activating factor X [159]. The normal FVILEVIa ratio in the serum is 100:1 (10 and 0.10 nmol/L, respectively), and with rFVI infusion, FVIIa lev- cls increase 100-fold to 3-20 nmoV/L [160]. The agent seems to be well tolerated and widespread activation of coagulation resulting in DIC has not ‘been reported. rFVII has proven to be useful in other off label uses such as controlling bleeding in trauma, obstetrical complications, and surgical patients with complex coagulation disorders (161, 162) TEVIL corrects vitamin-K dependent decreases of coagulation factors [163]. This has led to its use in liver patients with a depletion in vitamin-K dependent coagulation factors. FVII is able to effectively reverse prolonged prothrombin time (PT) in cirrhotic patients without active bleeding, [164], and Lisman et al, concluded that a single dose of fFVIla could lead to stable clot formation in cirrhotic patients [165]. Results have been mixed when using tFVIa in liver transplant patients, There are multiple case reports and series reporting efficacy of rEVIa at varying doses such as 100 [166], 80 (167], or 684 pg/kg [168]. However, better information is derived from several randomized control and retrospective case control studies. ‘Two retrospective case control studies con cluded that administration of ¢FVIa preopera- tively decreased blood transfusion requirements, in transplant patients [169, 170]. The dose of #EVIla in one of those studies of 22 patients was 58 pg/kg. Two randomized control studies failed to show efficacy of #FVIIa in reducing transfu- sion requirements while using 20, 40, 60, 80, oF 120 g/kg bolus doses of rFVila when com- pared to placebo (171, 172]. A more recent ran- domized, double-blind trial found that patients undergoing orthotopic liver transplant who received 40 g/kg of rFVila perioperatively required fewer transfused units of blood when compared to placebo [173]. A recent meta-anal- ysis on the use of rFVIla for bleeding prophy- laxis in hepatobiliary surgery was unable to draw conclusions for clinical application due to limited data [174]15 Coagulopathy: Pathophysiology, Evaluation, and Treatment 187 As we previously stated, prothrombin time does not necessarily correlate with improved ‘coagulation profile in liver disease, and comec- tion of prothrombin time does not necessarily decrease transfusion rate or improve outcome. ‘While it seems that more randomized, controlled studies should be implemented, the current con- sensus European Guidelines on the use or rF VIIa, during liver transplantation or resection recom- ‘mends that FVIla not be used routinely (grade B evidence) [175] Topical Agents in Hemosta: Several types of agents are available for intraop- ‘erative topical application to stimulate hemosta- sis, Products that provide a matrix for coagulation to occur such as collagen and cellulose, those that ‘mimic coagulation like fibrin sealants, and prod ucts that combine the use of endogenous and exogenous coagulation factors are currently employed during liver transplant surgery (31, 176]. Results are mixed showing that while there may be a decrease in blood transfusion intra and peri-operatvely, there may actually be no mortal~ ity benefit [177, 178]. A 2003 Cochrane review [179] and much ofthe endoscopic literature sup- ports the utility of fibrin sealants; however, aran- domized study of 300 liver resection patients found no difference in total blood loss, transfa- sion requirement, or morbidity with of without the produet 180} Summary Coagulation management is a fascinating and rapidly developing subfield in wansplantion. It extends across all fields of medicine presenting challenges for anesthesiologists, hepatologists, and liver transplant surgeons. Patients with liver disease are in a state of rebalanced hemostasis that may be affected by a number of factors. A high INR does not necessarily translate to increased bleeding risk in this patient population, ‘Viscoelastic testing may facilitate better assess- ‘ment of bleeding risk and coagulation balance prior to liver transplant and other invasive proce- dures. It is essential to optimize platelet count prior to surgery, however, unnecessary platelet and RBC transfusion should be avoided. Further studies are needed to improve our understanding, of the intricate imbalances comprising coagulop- athy in the liver disease patient, and to more effectively perform invasive procedures and care for the liver transplant patient, References 1. Davie EW, Ratnolf OD. Waterfall sequence forintrin- sicblood cating Science, 1968;145(3638) 1310-2. 2, Hoffman M, MonfoeDM, Robes HR. Cellular sctons in hemostasis, Haemostasis, 1996;26(Suppt D:D-6 3, Oliver JA, Monroe DM, Roberts HR, Holman 'M, Thrombin activates factor XT on activated plate- Jets inthe absonoe of factor XI. Arterioscler Thromb ‘Vase Biol. 1999:19(1):170-1, 4, Baglia FA, Walsh PN. Prothrombin is cofactor for the binding of fator XI to the platelet surface and fo platclet-mediated factor XI activation by thom- bin. Biochemistry. 1998:3(8)-2271-81 5, Franssen J, Salemink 1, Willems GM, Wun TC, Hemker HC, Lindhout T. Prothrombinase is pro- tected from inactivation by tissue factor pathway inhibitor: competition between prothrombin and ‘inhibitor Biochem J. 1997:323(Pt1):33-1. 6, Lu G, Broze GI Je, Krishnaswamy §, Formation of factors IXa and Xa by the extrinsic pathway fiferenial regulation by tissue factor “pathway ‘hibitor and. antithrombin II J. Biol Chem. 2004:279(17):17241-9. 7, Bsmon CT, Steaflo J, Suttie JWA. new vitain K-dependent protein. A _phospholipc-binding zymogen of a serine esterase. J Biol Chem. 1976:251(10):3082-6. 8. Eamon CT. Te endothelial protein C receptor. Cur Opin Hematol 2006;15(5):382- 9, Esmon CT, Esmon NL, Haris KW. Complex forma- tion between thromisin and thrombomodaln inhibits both thrombin-catalyzed fibrin formation and factor V activation. J Biol Chem, 1982;257(14):7944-7, 10, Lisman T, Leebsck FW. Hemostatic alterations in liver disease: a review on pathophysiology, clinical consequences, and treatment. Dig Surg. 2007;24(4):250-8, 11 Tripod A. The coagulopathy of chronic liver i east: is there a caus relationship with bleeding? No, Bur J Inter Med, 2010;21(2)65-9. 12, Lisman T, Bongers TN, Adelmeijer J, Janssen HL, de Maat MP, de Groot PG, e al. Elevated levels of vvon Willebrand Factor in cithosis support plate-