Bleeding is a common clinical presentation in small animal practice. It can be mild and self-limiting but also life-threatening requiring immediate medical attention. For the most part, hemorrhage is due to local vascular injury (e.g., caused by trauma, ulcers, tumors). Less frequently hemorrhage is caused by a disorder of primary or secondary hemostasis. Such disorders may cause abnormal clinical bleeding or only subclinical abnormalities on test results. Bleeding disorders occur more commonly in dogs than in cats.

Hemostasis is a complex interaction between platelets, blood vessels and plasma proteins. For didactic purposes, the hemostatic system is divided into 3 steps: 1) primary hemostasis, 2) secondary hemostasis, 3) fibrinolysis. This classical model is helpful to explain the main reactions and clinical manifestations of different hemostatic disorders and to understand the different hemostatic tests. Since there are close relationships between platelets, coagulation proteins and tissue factor bearing cells the classical model has been replaced by a "cell-based model of blood coagulation".

Primary hemostasis describes the processes of the primary closure of an injured blood vessel by the rapid formation of the unstable "primary platelet plug". It is dependent on an adequate platelet number and function, on adequate interaction between platelets and damaged vascular endothelium, and on the von Willebrand factor (vWF). vWF is a large protein consisting of multiple subunits synthesized by endothelial cells. It functions primarily as a glue, permitting adhesion of platelets to exposed collagen in the subendothelial matrix. In combination with vasoconstriction the primary platelet plug may be sufficient to stop capillary bleeding. For the closure of larger vessels the unstable plug has to be stabilized by insoluble fibrin.

The blood coagulation system (secondary hemostasis) results in the generation of thrombin. Thrombin converts soluble fibrinogen to insoluble fibrin. Classically, the coagulation cascade which is a series of enzymatic reactions involving coagulation factors has been divided into intrinsic and extrinsic pathways with a final common pathway. The intrinsic system is initiated by contact activation of FXII and the extrinsic pathway by the interaction of tissue factor and FVII. The cascade is useful when interpreting diagnostic hemostatic tests but does not reflect the in vivo situation since there is intensive interaction between the pathways. Contact activation is required for in vitro but does not play a major role for in vivo coagulation (FXII deficient animals do not bleed). Most coagulation factors and cofactors are synthesized in the liver and circulate in their inactive forms in the plasma. Factors II, VII, IX, X are considered vitamin K-dependent factors.

Natural inhibitors of blood coagulation (antithrombin, protein C pathway, tissue factor pathway inhibitor) which are primarily located on intact endothelial cells are essential for the regulation of blood coagulation. The final stage in hemostasis is the repair of the damaged vessel wall and dissolution of the fibrin clot. Fibrinolysis is mediated by the proteolytic enzyme plasmin. Plasmin degrades fibrinogen and (cross-linked) fibrin which results in the fibrin(ogen) split product (FSP) and D-dimer formation.

In cases of spontaneous, multifocal, and unexpected severe bleeding, a hemostatic disorder is suspected.

Hemorrhages caused by disorders of the primary hemostatic system are mainly related to platelets and rarely to vessels. Platelet disorders may be quantitative [thrombocytopenia due to (immune-mediated) accelerated destruction, bone marrow production defects, increased consumption, sequestration, or a combination thereof] or qualitative (thrombopathia). Spontaneous bleeding rarely occurs at platelet counts > 30,000/µl if there is no thrombo-, coagulo- or vasculopathy present in addition. Acquired thrombopathias are more common than inherited thrombopathias, but they rarely cause spontaneous bleeding. Causes include drugs (e.g., aspirin), uremia, dysproteinemia (e.g., in leishmaniosis, multiple myeloma), liver disease, or neoplasias. Inherited thrombopathias have been described in certain breeds (e.g., Glanzmann's thrombasthenia in Otterhounds). Von Willebrand disease (vWD) is often regarded as a form of thrombopathia. However, structure and function of platelets are normal. Increased tendency for bleeding is rarely caused by vasculopathies (e.g., vasculitis). Secondary hemostatic disorders can also be acquired or inherited (e.g., factor deficiencies such as hemophilia A and B). Causes of acquired coagulopathies include rodenticide intoxication, liver failure, vitamin K malabsorption, neoplasias). Some disease processes cause combined hemostatic defects (e.g., disseminated intravascular coagulopathy, DIC).

Diagnostic Approach

Signalment & history: A thorough history is important for the differentiation between inherited and acquired disorders. Breed, young age, similar problems in parents and/or siblings, earlier bleeding complications (e.g., after dentition, estrus, trauma, or surgery) are indicators of an inherited disorder. Information regarding vaccination status, drug exposure (e.g., aspirin, cytotoxic drugs), duration and clinical signs of hemorrhage, and known diseases (e.g., renal insufficiency, liver disease) are important. The possibility of exposure to toxins such as rodenticides should be investigated.

Clinical examination may help to differentiate between primary and secondary hemostatic defects. Moreover, the amount of blood loss can be estimated (color of mucous membranes, capillary refill time, and pulse quality). Dependent on the amount and the velocity of blood loss, either mild to severe anemia and/or hemorrhagic shock will result. Surface bleeding of the skin and mucosae such as petechiae, ecchymoses, epistaxis, gingival bleeding, melena, hematochezia, hematuria, retinal bleeding, scleral bleeding, or prolonged bleeding after injury are typical signs of a primary hemostatic disorder. Hematoma formation may occur. Characteristic signs of coagulopathies include bleeding into body cavities (e.g., hemothorax, hemomediastinum, hemoabdomen), into joints, hematomas, or delayed bleeding after injuries, but gastrointestinal hemorrhage and bruising may also occur.

Laboratory testing: Blood samples for investigation of hemostatic abnormalities have to be taken before starting treatment, since therapy (e.g., transfusions, vitamin K) can alter laboratory results and complicate the diagnosis. Atraumatic venipuncture with discarding of the first few drops of blood (to avoid platelet activation and tissue factor), appropriate sample handling and submission are very important.

Global tests assess the overall hemostatic function of a blood sample including the relationship between the platelet count and coagulation system. The whole blood clotting time in the serum tube is rather imprecise and not standardized. Automatic viscoelastic point-of-care hemostatic assays (thrombelastography, thrombelastometry) are particularly useful for the detection of hypercoagulable states. Generally, these methods have a limited sensitivity with respect to the detection of decreased individual hemostatic components.

A minimal database of a bleeding patient comprises a CBC including a platelet count, a clinical chemistry as well as selected coagulation tests. A blood smear is useful to provide a platelet estimate, to identify platelet size and clumping (and to examine for blood parasites, to evaluate RBC morphology, and to perform a differential blood count). Each platelet per high power oil emersion field represents approximately 15–20,000 platelets/µl blood. Obvious platelet clumps on the blood smear prevent accurate estimations. In cats incorrect low platelet values can occur with automated counting.

In cases of normal but also of decreased platelet counts, coagulation testing [activated partial thromboplastin time (APTT), prothrombin time (PT)] should be performed. Coagulation testing is needed to identify a defect in the intrinsic (APTT), extrinsic (PT) or the common pathway (APTT/PT) and to further investigate low platelet counts which might be caused by a complex hemostatic disorder such as DIC. Individual clotting factors have to be < 30% before PT or APTT are prolonged. Point of care coagulation analyzers (e.g., IDEXX Coag Dx) have been evaluated for dogs and cats. In cases of emergency the activated clotting time (ACT) can be determined. ACT tubes contain diatomaceous earth to act as contact activator for FXII (intrinsic system). In case of severe thrombocytopenia, ACT is prolonged due to a lack of platelet phospholipids. The clotting factors have to be < 10% of normal to prolong the ACT (dog reference value < 110 sec).

The thrombin time (TT) is a functional assay for fibrinogen to form fibrin and it is independent of vitamin K dependent coagulation factors. Therefore TT is useful to differentiate between rodenticide intoxication (normal TT, severely prolonged PT /APTT) and other complex coagulopathies (e.g., DIC, liver failure).

Laboratory abnormalities in cases of (severe) DIC include thrombocytopenia, schistocytes, APTT/ PT/ TT prolongation, positive FSPs, positive D-dimers, and decreased antithrombin concentration. In veterinary medicine, DIC is suspected if in addition to an underlying disease at least 3–4 abnormal test results are present.

If an inherited coagulopathy such as hemophilia A is suspected specific factor analyses can be performed in specialized laboratories.

If secondary hemostatic defects are excluded and the platelet count is normal or only slightly decreased (> 100,000/µl), a buccal mucosal bleeding time (BMBT) is performed. This is a simple in-vivo screening test of primary hemostasis. A prolonged BMBT is compatible with platelet dysfunction, vWF deficiency, or a vessel wall disorder (e.g., vasculitis). A coagulopathy will not immediately affect BMBT, but re-bleeding may occur. Platelet dysfunction is definitively diagnosed by using more comprehensive tests of platelet structure and activation response such as platelet aggregation tests. Functional platelet tests should be completed within 4 hours after blood collection; they require specialized and expensive equipment. "In-house" platelet function analyzers (PFA-100, Multiplate impedance aggregometer) have been evaluated in veterinary medicine.

Measurement of plasma protein, -urea, -creatinine, -bilirubin, liver enzymes, and an urinalysis serve to identify underlying diseases. Bilirubin can be increased in cases of internal bleeding. If gastrointestinal bleeding is suspected testing for occult blood should be performed.

Diagnostic imaging is necessary to diagnose bleeding into body cavities or organs such as the spleen. Moreover, they serve to evaluate potential causes of the hemostatic disorder (e.g., a neoplasia).

The therapeutic approach depends on the severity of bleeding, the existence of anemia and/or shock, and also on the underlying disease. If hypovolemia is present infusions of crystalloid fluids are indicated. Colloids should be used with caution since they might worsen a bleeding tendency. If in cases of acute blood loss the Hct falls below 0.20–0.25 l/l RBC transfusions or blood substitutes are indicated. The blood product used (packed RBCs, whole blood) depends on the availability and on the type of the hemostatic disorder.

Methods for minimizing hemorrhage include local hemostasis (wound pressure, ligations, topical agents), strict cage rest, gentle handling, and avoidance of elective surgery and other procedures such as intramuscular or subcutaneous injections. In addition, drugs that might interfere with hemostasis, such as NSAIDs, should be avoided.

In cases of hemorrhage due to severe thrombocytopenia (or thrombopathia) transfusions of platelets (e.g., fresh whole blood, platelet rich plasma) can be very effective. It is less beneficial in immune-mediated thrombocytopenia, where transfused platelets may be rapidly destroyed. Nonetheless, platelet transfusions can be life saving in cases of acute blood loss due to severe thrombocytopenia or if surgery is needed. Bleeding patients with vWD receive plasma products (fresh or fresh frozen plasma, cryoprecipitate) or fresh whole blood if severe anemia is present. In cases of mild bleeding or prior to surgery, a synthetic analogue of vasopressin (desmopressin acetate 1 µg/kg SC) can be injected 20 min pre-operatively.

If the hemorrhage is due to an inherited coagulopathy fresh (frozen) plasma is transfused. Transfusion of fresh whole blood is efficacious, but should be administered only in cases of severe anemia due to the risk of sensitization against RBCs.

If a vitamin K deficiency due to rodenticide poisoning is suspected, the therapeutic principles include: 1) induction of vomiting within the first hours of ingestion, 2) prevention of further hemorrhage and reversal of the coagulopathy by the application of vitamin K1 (4 mg/kg on day 1, then 2–3 mg/kg in divided doses for several weeks) and the transfusion of whole blood or fresh (frozen) plasma, and 3) correction of the anemia and of secondary signs such as dyspnea related to hemothorax. 1–2 days after discontinuation of vitamin K administration, PT has to be measured to indicate whether further treatment is required.

For patients bleeding due to DIC the prognosis is guarded or even poor. Management of DIC includes: 1) treatment of the underlying disorder, 2) maintenance of tissue perfusion/ improvement of microcirculation by appropriate infusion therapy, 3) prevention of secondary complications, 4) transfusion/ substitution therapy (whole blood, fresh or fresh frozen plasma), 5) heparin therapy (controversial).