Homeostatic regulation of coagulation is complex and heavily dependent on the liver. This process involves synthesis, activation and clearance of procoagulants, anticoagulants, and regulators of fibrinolysis, and hepatic synthesis of thrombopoietin, a glycoprotein hormone essential for thrombopoiesis ().
Because the liver is central to this homeostatic balance, hepatobiliary disorders complexly impact coagulation status. In liver disease, a continuum of readjustments rebalance or imbalance coagulation status. In some cases a decline in procoagulant activity is adjusted against a decline in anticoagulant activity, achieving copacetic changes.
Because most veterinarians in private first-opinion practices rely on traditional coagulation assessments (prothrombin time [PT] and activated partial thromboplastin time [aPTT]), and fibrinogen, with occasional measurement of antithrombin (AT), protein C (PC), D-dimers, and fibrin degradation products (FDPs), it is important to acknowledge the complexities that these may not reflect.
Of particular concern is hepatocyte-dependent vitamin K carboxylation of FII, FVII, FIX, FX, protein C, and protein S. Vitamin K activation of the dependent coagulation and anticoagulant proteins orchestrates carboxylation of glutamic acid residues, needed for reaction with calcium required for their hemostatic activity. The liver is the primary site of vitamin K storage; however, it only maintains short-term supplies, requiring intermittent frequent replenishment.
Vitamin K assimilation is dependent on an intact enterohepatic circulation of bile acids. Thus, its adequacy is thwarted in cholestatic disorders compromising enteric bile acid recycling (eg, complete extrahepatic bile duct obstruction [EHBDO], severe small bile duct ductopenia, and impaired bile flow or hepatic synthetic failure in cats with severe hepatic lipidosis).
Enteric availability of vitamin K also is compromised by illnesses causing hyporexia or anorexia and may be negatively impacted by chronic antimicrobial administration that may disrupt enteric flora involved with vitamin K precursor transformation.
Failed vitamin K factor activation also may simply reflect insufficient hepatic synthetic capacity. Complete major bile duct obstruction in cats of 7 days is long enough to provoke vitamin K–responsive coagulopathy. Cats with severe hepatic lipidosis can have vitamin K–responsive coagulopathy. Compared to dogs, cats with liver disease appear predisposed to vitamin K–responsive coagulopathies.
Based on trends in clinical practice, more dogs than cats undergo diagnostic liver biopsies, often because of unexplained chronic increases in liver enzyme activities. It has been estimated that up to ~64% of dogs with histologically documented liver disease that have undergone diagnostic abdominal ultrasonography have findings within normal limits. Thus, liver biopsy is often pursued to ascertain a definitive diagnosis to allow for rational interventional treatment, especially given the high incidence of copper storage hepatopathy in dogs of all sizes and breeds.
Clinical Assessments of Coagulation in Hepatic Disease in Small Animals
Clinical assessment of coagulation status of a patient with hepatobiliary disease begins with consideration of a thorough history. This includes notation of fecal color.
Acholic feces implicate potential vitamin K deficiency due to impaired vitamin absorption from interrupted enterohepatic bile acid circulation that accompanies major bile duct obstruction or severe small duct ductopenia. Evidence of enteric bleeding is queried as a manifestation of dysregulated hemostasis and spontaneous hemorrhage, disease-related abnormalities, or perhaps, secondary to splanchnic hypertension and acquired portosystemic shunts (APSSs).
Melanic stools (black, sticky feces) reflect high enteric bleeding (gastric or small intestinal sources) or hemobilia (gallbladder bleeding—necrotic cholecystitis, organ-specific vasculitis, mucosal trauma from choleliths, or neuroendocrine neoplasia).
Hematochezia (nondigested red blood in stool) implicates lower ileal or colonic bleeding, sometimes reflecting portal hypertensive vasculopathy.
Hematuria (blood in the urine) might reflect presence of ammonium biurate uroliths causing mechanical trauma to the urinary bladder or urethra or a rare acquired portosystemic shunt anastomosing with a ureter, urinary bladder, or kidney.
The owner should be queried about the possibility of observed spontaneous bleeding and the husbandry practices for control of endoparasitism. Fecal assessment for parasites should be ascertained.
Physical examination findings should be reconciled with historical information. Physical examination should include the following:
inspection of the mouth, pharynx, sclera, mucous membranes, and nonpigmented skin for petechia or ecchymotic hemorrhages and jaundice
visual scan for inappropriate bruising or prolonged bleeding from recent venipuncture or IV catheter sites
digital rectal examination to investigate stool color
examination of retinal microvasculature for evidence of hemorrhage
Platelet count should be assessed either by light microscope inspection of a blood smear. Platelet count can be estimated as follows: using a 100× oil immersion objective lens, determine the mean number of platelets visible in an ultrahigh power field (count at least 10 fields). The platelet count is calculated as the mean number of platelets per field multiplied by 15,000/mcL (eg, a mean of 2 platelets per ultrahigh power field corresponds to a platelet count of 30,000 platelets/mcL).
On blood smear assessment, platelet clumping on the feathered edge or throughout the smear should be considered because this will artifactually decrease the platelet count/mcL.
Thrombocytopenia < 30,000 platelets/mcL is associated with high risk for iatrogenic hemorrhage during liver biopsy collection.
Abnormalities compromising primary hemostasis (initiation and formation of a platelet plug) include the following:
inadequate platelet number (thrombocytopenia, < 30,000 platelets/mcL)
platelet dysfunction (thrombocytopathia)
vascular disease (vasculitis)
alterations in function or quantity of VWF/FVIII interaction
Some of these derangements may have a disproportionate impact on the liver such that assessment made on blood collected from the systemic circulation may give a false sense of well-being.
If the platelet count is > 30,000 platelets/mcL, measurement of a buccal mucosal bleeding time (BMBT) is indicated if a liver biopsy is being considered or bleeding tendencies are appreciated. The mucosal side of an inverted lip (position secured with a loose gauge tie around the muzzle, best if the dog is sedated), free of congested compression, is used for this assessment. A spring-loaded automated lancet or template device is used to produce a standard depth and length incision. Blood from the undisturbed incision is gently blotted from below the wound, using absorbable paper every 5 seconds until the time until cessation of bleeding is ascertained.
A BMBT is reported as normal (< 5 minutes) or abnormal.
If bleeding persists > 15 minutes or if rebleeding occurs, tissue glue is used to achieve hemostasis. There is no direct correlation between the duration of abnormal bleeding of a BMBT and the severity or grade of dysregulated coagulation.
The sensitivity and relevance of BMBT assessments in predicting spontaneous or iatrogenic biopsy-related bleeding remains unsubstantiated in veterinary medicine. However, the lack of sensitivity of the PT and aPTT to detect risk for bleeding has popularized the BMBT, with some clinicians always making this assessment just before securing a liver biopsy. Others use this test when PT and aPTT results are slightly increased and a liver biopsy is impending.
If the BMBT is normal, the procedure can be completed, likely without hemorrhagic complication. If the BMBT is abnormal and a liver biopsy essential, plasma transfusion and desmopressin can be administered and the BMBT reevaluated. If within normal limits, the liver biopsy is collected. If the BMBT remains abnormal, the owner should be informed of the procedural risk for bleeding.
Next, coagulation status beyond primary hemostasis should be assessed. Secondary hemostasis, as illustrated in , involves a complex series of interactive proteases that are consecutively activated, with an ultimate endpoint of balanced thrombin generation and inhibition. Secondary hemostasis should reinforce and modulate the primary hemostatic response.
Complex alterations involving secondary hemostasis in patients with hepatobiliary disease may variably involve the abnormal activity (abundance, insufficiency, dysfunction) of procoagulant factors (FI, FII, FV, FVII–FXIII) and anticoagulant factors (AT, PT), including adequacy and strength of formed fibrin. Evaluation of the PT or aPTT clotting times only provides a superficial snapshot of a complex and dynamic series of interactions.
In some animals with hepatobiliary disease, a tenuous hemostatic balance is achieved, disproportionately impacting the liver. Disrupted sinusoidal vasculature (caused by the underlying liver disorder) increases local release of VWF-FVIII complexes, escalating platelet aggregation on damaged endothelium. This process interrupts sinusoidal perfusion that may be aggravated by formation of neutrophilic extracellular traps (NETs), which are composed of DNA-histone complexes and proteins released by activated neutrophils. NETs play an important role in promoting thrombosis; increased local activation of procoagulants, anticoagulants, and fibrinolytics lead to microvascular thrombi.
While often restricted to hepatic sinusoidal vasculature, disrupted coagulation may extend to influence the systemic circulation. In dogs with necroinflammatory liver disease (hepatitis), a rare sequela to unbalance hemostasis is thrombosis of the portal vein. Portal vein thrombosis (PVT) reflects the hypercoagulable status of the patient, sluggish transhepatic perfusion caused by sinusoidal collapse, and disease-related liver remodeling and fibrosis.
Fibrinogen in Hepatic Disease in Small Animals
Studies in humans and observational data in dogs deem fibrinogen concentration an influential determinant of bleeding and thrombotic risk in hepatic disease. Concentrations of fibrinogen generally decline substantially only in advanced-stage liver disease, acute fulminant liver failure, or severe acute-on-chronic liver injury. These patients exhibit increased risk for bleeding. Synthetic failure may initially be camouflaged by the fibrinogen acute-phase protein response.
In the absence of a functionally measured fibrinogen concentration, a crude assessment of fibrinogen can be deduced from the difference in total solids concentration between samples of plasma and serum, with most of the difference reflecting fibrinogen. However, low fibrinogen concentrations also may derive from consumptive coagulopathy. In that case, D-dimers will also be markedly increased. Unfortunately, D-dimers are frequently increased in patients with clinical liver disease, at least because of their decreased clearance, but also likely because of increased generation.
Fibrinogen concentration < 100 mg/dL in dogs and < 70 mg/dL in cats usually reflects synthetic failure or consumption coagulopathy.
Increased fibrinogen concentrations, provoked by sinusoidal inflammation and endotoxemia, add to the thrombotic imbalance of a hypercoagulable status. In humans, markedly increased fibrinogen is associated with risk for PVT, usually as a consequence of chronic liver disease and portal vein hypertension. PVT is not a common sequela of chronic liver disease in the dog or cat although it can occur.
Other disease processes provoking a hypercoagulable status also can provoke PVT (eg, nephrotic syndrome, severe protein-losing enteropathy, severe eosinophilic enteritis, immune-mediated hemolysis anemia, hyperadrenocorticism, chronic glucocorticoid administration, various types of neoplasia, and severe thrombocytosis).
Measurement of fibrinogen is a useful metric in estimating the risk for bleeding versus thrombosis in animals with hepatobiliary disease. Understanding that development of a consumptive coagulopathy may mask directional trends is essential in deducing the presence of hepatic synthetic failure.
Thrombocytopenia and Altered Platelet Aggregation in Hepatic Disease in Small Animals
Mild thrombocytopenia (platelet count > 80,000 platelet/mcL) is recognized in some dogs with hepatic disease. This may reflect splenic sequestration of platelets secondary to splanchnic hypertension, impaired thrombopoiesis because of decreased hepatic synthesis of thrombopoietin, or consumptive utilization. Decreased synthesis of thrombopoietin by the liver is considered a major cause of thrombocytopenia in humans with advanced-stage liver disease, rather than splenic sequestration (secondary to portal hypertension).
Low-grade consumptive utilization of platelets would fit the paradigm associated with the marked increase in VWF. Overt disseminated consumptive coagulopathy usually is associated with one or more additional classic indicators, such as circulating schistocytes, abnormally prolonged or rapid PT, aPTT, extremely high or low fibrinogen, increased D-dimers or FDP, and subnormal anticoagulants (PC or AT).
Liver disease in humans can impact platelet function where hypoaggregability is more common than hyperaggregability. Limited studies in dogs corroborate similar acquired platelet dysfunction with some liver disorders. Altered platelet functionality can occur with or without bleeding tendencies and without abnormal coagulation metrics (ie, PT, aPTT, or BMBT).
Decreased platelet aggregation was documented in dogs with portosystemic vascular anomaly (PSVA) without association with bleeding tendencies. Two studies examining platelet aggregation in dogs with chronic hepatopathies documented decreased platelet aggregation in a subset of dogs, many without abnormal coagulation metrics or bleeding tendencies. Numerous human studies document similar findings; however, with newer aggregation technology, it has been recognized that subsets of platelets can display hyperaggregability that could foster development of thrombi. Platelet functional assessments for dogs and cats are not available outside of large specialty or teaching hospitals.
Platelet aggregation is most reliably done on whole blood using impedance aggregometry. Platelet concentrated plasma should not be used because of unavoidable changes in the lipid interface affecting platelet membranes, inadvertent elimination of a subpopulation of relevant platelets, and sample agitation that may lead to platelet activation. A PFA-100 cartridge method is not ideal for assessment of platelet dysfunction in liver patients as this technique appraises platelets in a high shear stress microenvironment (arterial type flow).
The following are potential causes of platelet hypoaggregability in chronic hepatic disease:
altered thromboxane A2 or prostaglandin synthesis
altered membrane signal transduction
decreased expression of platelet receptors needed for the aggregation response
abnormal circulating lipoprotein profiles
circulating platelet inhibitors (nitric oxide, prostacyclin, bile acids)
Regardless of cause, the clinical importance of altered platelet function in humans with liver disease is still debated among hematologists and hepatologists. Even with abnormalities implicating dysfunction of primary hemostasis, associated bleeding has generally not been recognized. Potential causes of platelet hyperaggregability link with changes in VWF within the sinusoidal milieu and its molecular binding partner (Adam13).
von Willebrand factor is a large multimeric glycoprotein, considered an acute phase reactant that importantly contributes to primary hemostasis. Plasma concentrations of VWF increase in response to physical stresses and local and systemic inflammation. Activation of VWF on damaged vasculature functions as a platelet adhesion molecule, initiating platelet plug formation where a clot will ultimately develop. VWF functions as the carrier protein for FVIII, protecting this protease against premature activation, and is synthesized by megakaryocytes and endothelial cells.
Endothelial cell origin is important to the liver, where it is in part constitutively secreted; it has a half-life of ~12 hours, with the liver being its primary site of clearance. The remainder of VWF in the endothelial cell is stored in specific organelles (Weibel-Palade bodies). Megakaryocytes store ~15% of VWF packaged in alpha-granules of platelets. Large VWF multimers stored in endothelial cells and platelets orchestrate platelet vascular adherence.
Studies of hemostasis in liver disease (humans, animal models) clarify preservation and increased concentrations of FVIII while all other procoagulants decline. This phenomenon is thought to reflect extrahepatic FVIII synthesis induced by liver injury; increased intrahepatic (sinusoidal endothelium) FVIII synthesis because endothelial cell functionality is retained against the landscape of hepatocyte injury; or endothelial cell activation by endotoxin (common in severe, acute, and acute-on-chronic liver injury). It is also possible that FVIII is salvaged by its protective affiliation with VWF; induced VWF concentrations in liver disease due to local inflammation exceed VWF clearance.
Increased VWF in liver disease is thought to auto-compensate for decreased number and function of platelets, in part explaining the lack of bleeding tendencies in patients with thrombocytopenia and decreased platelet aggregation. A 5- to 10-fold increase in VWF antigen and activity is documented in humans with acute and chronic liver disease. In fact, escalated VWF activity is shown to be an independent predictor of hepatic decompensation in patients with acute and acute-on-chronic liver disease and as a predictor of death in advanced-stage liver disease.
Some data in dogs and cats with liver disease document similar increases in plasma VWF activity. Studies in humans with liver disease suggest that increased VWF activity in cirrhosis compensates for abnormalities of platelet number or function. Increased VWF concentrations in hepatobiliary disease may reflect sinusoidal endothelial damage, enhanced endothelial surface area exposure, and increased circulating vasoconstrictors (such as vasopressin) as well as decreased hepatic VWF clearance.
While thrombocytopenia and decreased platelet function should provoke a hypocoagulable status in liver disease, the up to 10-fold greater increase in VWF in the liver in advanced-stage liver disease seemingly helps compensate hemostatic balance towards a normocoagulable status.
von Willebrand's testing in dogs with liver disease up to now has typically targeted identification of breed-related deficiencies, increasing risk for bleeding. Based on the current knowledge of VWF in liver disease, this metric should be used routinely in the hemostatic assessment of liver patients as a marker of imbalanced coagulability or compensated stabilization. In the presence of a notably increased VWF, the use of cryoprecipitate should be carefully considered because this might provoke a thrombotic complication.
Secondary Hemostasis Assessments in Hepatic Disease in Small Animals
As early as 1981, it was noted that hemorrhagic complications of liver biopsy in hepatic disease did not correlate with PT and aPTT metrics (human work). In these pioneering studies, clinicians observed that prolongation of PT or aPTT did not correlate with bleeding tendencies until values were > 1.5× the upper limit of the reference range. This landmark remains quoted to this day.
Deficiency of a single coagulation protein is not detected by the PT or aPTT assays until factor concentration is < 25%–30% of normal. If multiple coagulation factors are decreased to this level, they synergistically prolong PT and aPTT, as observed in liver disease.
Conventional coagulation assessments may not reflect altered coagulation balance in patients with hepatobiliary disease. This disconnect between projected risks and real-time case outcome has led many clinicians to abandon PT and aPTT assessments in patients lacking historical or physical evidence of bleeding tendencies. Rather, the more relevant assessment of primary hemostasis using the buccal mucosal bleeding time (BMBT), has been adopted (see BMBT discussion Clinical Assessments of Coagulation ).
With this approach, clinicians prepare for blood component and desmopressin-assisted crisis management. Assessment of PT and aPTT has relevant predictive value in suspected hepatic failure, where critical prolongation (> 1.5× the upper reference limit) warrants a grave prognosis (coagulation metrics considered in the context of clinical status and its progression).
The alternative to classic PT and aPTT assessments along with measurement of fibrinogen, D-dimers, and anticoagulant proteins (AT, PC) is thromboelastography. This technology provides a global snapshot of a patient’s circulating blood coagulability. Thromboelastography has been adopted into human medicine to estimate concomitant changes in procoagulant, anticoagulant, and fibrinolytic mechanisms in liver patients where “rebalanced hemostasis” is an ongoing process.
Coagulability predictions can be used to individually tailor therapeutic interventions. Thromboelastography has predicted bleeding tendencies in humans with advanced-stage liver disease undergoing liver transplantation.
However, despite its expanding use, there remains a chasm between thromboelastography-predicted coagulability status and the propriety and effectiveness of proposed blood component therapy and administration of anticoagulants and fibrinolytics. Outcomes of patients in several studies have not changed, although more rigorous scrutiny of patient status has led to more thoughtful recommendations for blood component therapy. A recent review of coagulation assessments in humans with liver dysfunction concluded that, overall, the body of evidence supporting thromboelastography as a guide for blood component therapy during liver transplant (the most challenging of all liver conditions) remains poor, despite its adoption in large academic transplant centers.
Thromboelastography provides information regarding: speed of clot formation, clot strength, clot stability, and fibrinolysis—thus a global functional snapshot of hemostatic balance between procoagulants and their inhibitors, formation and strength of fibrin strands, and the appropriateness of clot dissolution or fibrinolysis. Thromboelastography metrics include R (initial fibrin formation), K (clot formation time), angle (reflecting the rapidity of fibrin cross-linking), MA (indicative of overall clot firmness), and LY30 (expressing percentage of clot lysis during 30-minute period after MA is achieved). The G value represents a mathematical manipulation of MA and is used to categorically rank coagulability into normocoagulable, hypocoagulable, and hypercoagulable categories.
This methodology requires dedicated technical support to standardize the testing protocol and equipment maintenance essential for valid test results. An important consideration in thromboelastography testing is that anemia can erroneously skew findings toward interpretation as a hypercoagulable status. Additionally, thromboelastography does not evaluate the endothelial contribution to coagulation (important in liver disease), has low sensitivity to mild coagulation factor deficiencies or mild defects in primary hemostasis, and operationally, has variability and poor result reproducibility with the same operator and analyzer.
It is recognized that humans with advanced-stage liver injury achieve self-adjusting balance between procoagulants, anticoagulants, and fibrinolysis that measure in pathological ranges. However, these patients are intrinsically balanced in a manner that controls pathological hemorrhage. Thus, the prudence of making therapeutic recommendations or deducing prognoses based on thromboelastography findings is yet to be proven in large populations of dogs and cats. Simplistic interpretation of findings can lead to aggressive interventions.
Studies corroborate the complexity of hemostatic disruptions among dogs (fewer feline studies) with different hepatobiliary disorders. Indeed, findings for some disorders contradict long-held assumptions regarding coagulation status and elucidate risks heretofore not considered. Some animals with liver disease have coagulopathies favoring hypercoagulative complications (ie, risk of clot formation) despite traditional testing that suggests hypocoagulative or normal status.
The diversity among hemostatic classifications based on thromboelastography metrics coordinates with the discordance recognized between traditional coagulation assessments and experiential observations in patients with hepatobiliary disorders.
Interventional Options For Coagulation Dysregulation in Liver Disease
The tables and summarize the text that follows.
Administration of vitamin K is recommended for all patients with jaundice considered due to hepatobiliary disease on entrance to the hospital because of the poor sensitivity of the PT and aPTT for detection of vitamin K deficiency. This is extremely important in cats, where extrahepatic bile duct obstruction Extrahepatic Bile Duct Obstruction in Small Animals Extrahepatic bile duct obstruction is the blockage of the normal flow of bile from the liver to the intestinal tract. Obstruction of the common bile duct is associated with a number of diverse... read more (EHBDO) and HL can induce vitamin K insufficiency within 7 days of inappetence and jaundice.
Although dogs are at lesser risk for acute-onset vitamin K adequacy, following this protocol avoids delays in planning liver biopsy or exploratory surgery if these are recommended diagnostics.
Avoid IV administration, which may cause anaphylactoid reaction. Excess administration (ie, daily) in cats can cause Heinz body hemolysis.
Desmopressin mobilizes preformed large VWF monomers from endothelial stores as a one-time response phenomenon. Numerous studies in humans with advanced-stage liver disease suggest it can interrupt bleeding associated with liver biopsy but not enteric hemorrhage associated with portal hypertension. However, there is no general consensus on its recommended use.
Onset of action is within 1 hour of dosing with reports of duration of action up to 8 hours. Repeated dosing within 8 hours has no additive effect.
Desmopressin is used in animals undergoing liver biopsy if excessive bleeding is witnessed during the procedure or anticipated by BMBT. Too-frequent administration of desmopressin can provoke water retention and hyponatremia, with potential to worsen ascites. Intermittent single-dose administration does not pose this hazard.
Epsilon-aminocaproic acid, an antifibrinolytic, is used to control hyperfibrinolysis by reversible inhibition of plasminogen, plasmin-mediated binding to fibrin, and to a lesser extent, increased antiplasmin activity. This drug also affects platelet function by preventing plasmin-mediated degradation of glycoprotein 1b receptors. The effect of drug administration can be monitored with thromboelastography assays set for assessment of clot lysis.
The other drug used to control hyperfibrinolysis is tranexamic acid, which has a more potent effect and greater likelihood of iatrogenic bleeding. There is no experience using this drug for animals with liver disease.
The use of NSAID-type platelet inhibitors is strongly discouraged in patients with liver disease because of potential for hepatotoxicity. Rather, thienopyridines (eg, clopidogrel) are recommended. Drugs in this category are direct platelet agonist receptor inhibitors and antagonists to ADP platelet receptor (P2Y12).
Because these anticoagulants are pro-drugs requiring liver activation (via cytochrome P450-dependent oxidation), it is possible that animals with advanced-stage liver disease may not achieve expected platelet inhibition or have other pharmacokinetic alterations requiring tailored drug dosing. The active metabolite irreversibly binds the ADP receptor, impairing the release of serotonin, thromboxane, and ADP and the inhibition of ADP-mediated activation of the glycoprotein (GP) IIb/IIIa receptor.
Low molecular weight heparin (LMWH) is derived from depolymerization of unfractionated heparin and is less protein-bound with more predictable pharmacokinetic profiles and better bioavailability after subcutaneous injection. The smaller size of the LMWH polysaccharides limits their simultaneous binding of AT and thrombin; thus they have decreased inhibition of FIIa relative to anti-Xa compared to unfractionated heparin (~one-third of unfractionated heparin FIIa inhibition).
The LMWHs provide a more predictable anticoagulative effect that can decrease need for frequent therapeutic monitoring. However, monitoring is still needed to achieve tailored optimal dosing for that patient. Specific monitoring for anti-Xa activity improves therapeutic dosing.
Rivaroxaban is a drug engineered to directly inhibit FXa and can be orally administered. It improves client compliance and decreases the need for frequent monitoring. Assessment of PT coagulation time has been shown to have good correlation with a drug-specific assessment method.
Blood Component Therapy in Hepatic Disease in Small Animals
Finding PT or aPTT > 1.5× the upper reference interval in an animal not demonstrating evidence of bleeding, when the patient is not scheduled for surgical procedures, is not a mandate for blood component administration in hepatic disease. Blood component therapy is costly, does not achieve a longterm corrective outcome in liver disease, and may be associated with adverse effects.
It is important to consider the following:
Continual homeostatic self-adjustment occurs in patients with liver disease that can balance hemorrhagic and coagulopathic propensities.
Blood component intervention does correct these instabilities without modifying the underlying disease process.
Changes achieved with blood component infusions are transient and are assessed based on patient response (ie, bleeding is curtailed) rather than measured metrics.
Blood component therapy is advised when a patient with hepatobiliary disease is bleeding and has abnormal hemostatic metrics, with demonstrated failure to self-correct. This usually applies to patients with acute severe hepatic failure that needs to be bridged until regenerative recovery. In acute liver failure due to some hepatotoxicoses, this may not be possible.
Acute-on-chronic liver injury, as occurs in advanced-stage liver disease, usually is associated with ascites, acquired portosystemic shunts (APSSs), and enteric bleeding associated with a splanchnic vasculopathy. Often these patients will not improve with heroic massive blood component therapy or surgical interventions to ligate or embolize collateral shunting vasculature. Owners need to be informed of the dismal outcome prospects before such therapeutic endeavors are initiated.
Platelet counts < 30,000 platelets/mcL are considered a certain risk for iatrogenic hemorrhage as a stand-alone liability and warrant prompt canceling of invasive sampling or surgical procedures. Blood smears should be reassessed to validate the reported platelet count. If in question, resubmit a new blood sample and repeat automated platelet count with review of a blood smear looking for substantial platelet clumping (routinely encountered in cats).
Differential diagnoses for thrombocytopenia should be considered because hepatobiliary disease alone usually is not associated with this severity of thrombocytopenia. Considered causes should include tickborne infectious disorders, systemic or primary hepatobiliary neoplasia, and immune-mediated hematologic disorders.
A blood smear should be examined for the following:
abnormal RBC morphology (eg, spherocytes, schistocytes)
disseminated neoplasia (eg, lymphocytic leukemia)
pathological platelet clumping (pathological hyperaggregation)
In some cases, a bone marrow aspirate should be examined to define if thrombocytopenia is caused by peripheral destruction, bone marrow neoplastic invasion, myelofibrosis, or failed thrombopoiesis. Thrombocytopenic dogs rarely encounter serious hemorrhage from bone marrow sampling. Platelet transfusions are not recommended as a longterm strategy to manage thrombocytopenia associated with liver disease because of their short duration effect, inability to correct pathological thrombocytopenia, and expense. Additionally, platelet and microparticle aggregation in hepatic sinusoids and phagocytized by hepatic Kupffer cells may contribute to the sinusoidal inflammatory milieu and a prothrombotic status.
Special considerations are required for patients with acute severe liver injury and severe acute-on-chronic liver failure. Pathophysiologic details in these categorical disorders are associated with increased release of VWF from sinusoidal endothelium, especially when complicated by endotoxemia or systemic inflammatory response syndrome (SIRS). The acute-on-chronic proinflammatory milieu provokes endothelial activation with release of inflammatory mediators.
Endotoxemia may derive from system infection or inflammation or from permissive translocation of endotoxin from the alimentary canal. Increased exposure of hepatic cells to materials transported in the portal vein may occur secondary to damaged sinusoidal endothelium, impaired Kupffer cell phagocytosis, or gut inflammation. In one study of humans in decompensated liver failure, high VWF activity measured at presentation predicted a poor prognosis. In these disorders, VWF antigen and activity appreciably exceed those associated with stable cirrhosis. Coagulative support of these patients probably should not include cryoprecipitate because of its high concentrations of VWF.
Administration of whole or packed red blood cell (PRBC) products to patients with advanced liver disease, animals with APSSs, and some animals with severe acute liver injury and severe acute-on-chronic liver disease can provoke hepatic encephalopathy (HE) because of the realized hemoglobin challenge. Hemoglobin is the most hepatoencephalogenic protein source known; the pathogenesis of this association has not been elucidated. Hemoglobin proteinaceous waste challenges ammonia detoxification. Because PRBC and stored whole RBC contain more senescent erythrocytes, there is greater risk for hyperammonemia and HE.
Blood Component Therapy Considerations
Fresh whole blood is an ideal recommendation for correction of critical anemia in a patient with liver disease complicated by the need for coagulation support. Fresh whole blood delivers RBCs, platelets, coagulation factors, anticoagulant proteins, and VWF (small amounts). What qualifies as critical anemia is based on vital signs (tachycardia, weakness, cardiac arrhythmias) and usually a rapid decline of PCV to < 25%.
Packed red blood cells (PRBCs) pose greater risk for HE in at-risk patients. Decreased circulating life span of RBCs (relative to whole fresh blood) increases proteinaceous wastes challenging ammonia detoxification. Ensure that RBC mass is an essential therapeutic need if PRBCs are administered and there is no option for whole fresh blood. PRBCs provide no functional platelets, coagulation proteins, or albumin for a patient with advanced-stage liver disease, acute liver failure, or severe acute-on-chronic liver disease. As above, anemia qualifying as critical is based on vital signs (tachycardia, weakness, cardiac arrhythmias) and usually a rapid decline of PCV to < 25%.
See concerns about stored whole blood noted in . Stored whole blood contains coagulation proteins (may be nonactive), albumin, other proteins, and nonviable platelets. The decreased circulating lifespan of RBCs contributes to risk of HE, as above. Critical anemia is as defined for fresh whole blood.
Fresh frozen plasma is a universal source of circulating coagulation factors, albumin, globulins, antithrombin, and some VWF. It is the first choice for patients with notably prolonged PT and/or aPTT exhibiting bleeding tendencies or iatrogenic bleeding during liver biopsy.
Cryoprecipitate is a concentrated source of plasma VWF, FVIII, FXIII, and fibrinogen. Cryoprecipitate might provoke worsening hemostatic imbalance in dogs with hyperfibrinogenemia and high VWF activity.
Cryoprecipitate-poor plasma (plasma cryoprecipitate decreased) is plasma remaining after extraction of cryoprecipitate. It can deliver FII, FVII, FIX, FX, albumin, anticoagulants, and fibrinolytic factors but is deficient in VWF and FVIII. It may be the ideal blood component for patients needing coagulation factor support with high VWF that may be aggravating sinusoidal inflammation and hyperfibrinogenemia.
Platelet-rich plasma is indicated for rescue of uncontrolled life-threatening bleeding due to severe thrombocytopenia or thrombocytopathia, extremely uncommon in animals under consideration in this chapter. Platelet-rich plasma is expensive, the platelets have a short survival after transfusion (hours), and this treatment should be used as a short-term bridge coupled with another definitive step targeting disease management, as in immune-mediated thrombocytopenic dogs undergoing splenectomy. Platelet infusion might contribute to a hypercoagulable status, particularly at the sinusoidal level, in some patients with severe hepatobiliary disease.
Lyophilized platelets are available for veterinary use; however, there is no indication for administration of this product in the context of animal health discussed in this chapter. Furthermore, transfusion of a cryopreserved platelet product in thrombocytopenic dogs only transiently increased platelet count but failed to improve clinical bleeding or survival compared to a control group.