Treatment of Afibrinogenemia in a Chihuahua

Introduction

Three categories of fibrinogen disorders have been described in human and veterinary medicine. Afibrinogenemia is the lack of measurable levels of fibrinogen in the blood; hypofibrinogenemia is a decreased, but measurable, level of fibrinogen; and dysfibrinogenemia is defined as normal to decreased blood levels of either abnormally constructed or abnormally functioning fibrinogen. In humans, three genes are known to code for fibrinogen. Fibrinogen genes α, β, and γ are found on chromosome four. Congenital afibrinogenemia has an autosomal recessive inheritance and therefore requires either homozygosity or compound heterozygosity.¹ Several different mechanisms for afibrinogenemia have been described and can act at the DNA level, RNA level, or protein level. Hypofibrinogenemia also has an autosomal recessive inheritance, but requires heterozygosity affecting protein synthesis, assembly, or secretion.¹ Dysfibrinogenemia is inherited in an autosomal dominant mode.¹ Individuals affected with dysfibrinogenemia are heterozygous for a missense mutation at one of the three fibrinogen genes, causing abnormal fibrinogen function.

Patients with afibrinogenemia present with persistent bleeding and increases in both prothrombin time (PT) and activated partial thromboplastin time (aPTT), which could be confused for vitamin K rodenticide toxicity, hepatic failure, or disseminated intravascular coagulation (DIC). The purpose of this report is to discuss the diagnosis and treatment of afibrinogenemia in a Chihuahua.

Case Report

A 3 yr old female Chihuahua weighing 2.2 kg presented to the veterinarian for lameness of the right hind limb subsequent to falling from the owner’s bed. On physical examination, bruises were found on the medial aspect of the right hind limb and the dorsal aspect of the head. Radiographs of the thorax, abdomen, and right hind limb were within normal limits. Two blood samples were collected. One was for in-house evaluation of PT and aPTT^a and the other for submission to a reference laboratory^b for evaluation of PT, aPTT, fibrinogen, D-dimer, and platelet count. In-house coagulation times (PT and aPTT) exceeded the upper limits of detection. The veterinarian suspected vitamin K antagonist rodenticide toxicity and administered both a fresh frozen plasma transfusion^c (40 mL) and 10 mg vitamin K₁^d subcutaneously. The patient was discharged and prescribed vitamin K₁^e (10 mg per os q 24 hr for 3 wk). The reference laboratory results returned the next day showing a PT > 100 sec (reference range, 6.0–12.0 sec), aPTT > 100 sec (reference range, 10–25 sec), fibrinogen (performed by the Clauss method) below the lowest detection limit of the assay (< 50 mg/dL; reference range, 150–400 mg/dL), a normal D-dimer, and a platelet count of 527 × 10³/μL (reference range, 170–400 × 10³/μL).

The patient returned to the veterinarian’s hospital 24 hr after initial presentation for reevaluation. At that time, the PT and aPTT levels still exceeded the upper limits of detection. The patient was hospitalized for 48 hr and was administered a total of 165 mL (37.5 mL/kg/day for 2 days) of fresh frozen plasma before being discharged to continue oral vitamin K₁ therapy. A blood sample sent to a reference laboratory 3 days later had normal PT and aPTT levels but a prolonged thrombin clotting time (TCT) of 17 sec (reference range, 5–9 sec).

The patient presented to the veterinarian 2 mo later with a large bruise found on the abdomen, hematochezia, and vomitus containing small amounts of fresh blood. An in-house complete blood cell count (CBC) and biochemical panel revealed a moderate anemia with a hematocrit of 21% (reference range, 37–55%), a mature neutrophilia (31.7 × 10³/mm³; reference range, 6–17 × 10³/mm³) and mild hypoalbuminemia (2.1 g/dL; reference range, 2.3–4.0 g/dL). One unit of fresh frozen plasma (120 mL) was administered.

After another 2 mo, the patient presented to the referring veterinarian for neck pain and decreased appetite. A large hematoma was found on the ventral aspect of the neck over the external jugular vein. Radiographs of the thorax and abdomen were within normal limits. A CBC and biochemical panel were completed in-house and revealed a normal chemistry profile, a hematocrit of 35%, decreased hemoglobin (11.6 g/dL; reference range, 12–18 g/dL), and mild increase in the platelet count (550 × 10⁹/L; reference range, 175–500 × 10⁹/L).

The next day, the patient was referred to a board-certified internist for an undetermined coagulopathy. Radiographs were within normal limits, and a fresh frozen plasma transfusion (48 mL) was administered over 4 hr. The CBC revealed a neutrophilia with a left shift (segmented neutrophils were 18,249/μL; reference range, 3,000–11,500/μL), band neutrophils (1,185/μL; reference range, 0–300/μL), a nonregenerative anemia (hematocrit was 14.2%; reference range, 37–55%), and a slightly decreased platelet count (140 × 10⁹/L; reference range, 164–510 × 10⁹/L). The coagulation panel revealed a PT > 100 sec (reference range, 5.9–8.3 sec), an aPTT > 100 sec (reference range, 10.6–20 sec), and a fibrinogen level < 20 mg/dL (reference range, 90–255 mg/dL). Pre- and postprandial serum bile acids were within normal limits. An ultrasound of the abdomen was normal. A second fresh frozen plasma transfusion (66 mL) was administered based on the prolonged PT and aPTT. Coagulation times the following day were found to be within normal limits. A blood sample was collected (0.1 mL citrate/mL of blood) from both the patient and a normal canine control 6 days later. The blood samples were submitted for PT, aPTT, TCT, fibrinogen, antithrombin, D-dimer, and clotting factors II, VII, VIII, IX, X, XI, XII.^f That coagulation panel revealed a PT > 60 sec (reference range, 13–18 sec), an aPTT > 60 sec (reference range, 10–17 sec), a TCT > 60 sec (reference range, 5–9 sec), a mixed TCT (performed by diluting the patient’s plasma 1:1 with the plasma from control) within normal limits, a fibrinogen level (assayed via the Clauss method) < 15 mg/dL (reference range, 147–479 mg/dL), a and D-dimer of 250–500 ng/mL (reference range, < 250 ng/mL). Antithrombin, factor II, and factors VII–XII were all within normal limits. All of the control samples were within normal limits ruling out both sampling and laboratory error. To exclude previously reported antibodies against fibrinogen, samples of the patient’s plasma were diluted 1:3, 1:1, and 3:1 with the control patient’s plasma.² Fibrinogen levels were calculated for a dilution effect and compared with the measured fibrinogen levels. All fibrinogen readings were consistent with the predicted dilution values indicating that fibrinogen antibodies were unlikely.

Two days after discharge from the referral hospital with a diagnosis of hypoalbuminemia, the patient had vomiting and diarrhea secondary to dietary indiscretion. The patient’s vomiting and diarrhea progressed to include hematemesis, hematochezia, and melena. The patient was taken to an emergency hospital and was treated with a 60 mL whole blood transfusion. Posttransfusion, the PT and aPTT were within normal limits (determined using an in-house analyzer), and the patient was transferred back to the referral hospital the following day. The patient continued to bleed from the venipuncture sites and was vomiting fresh and digested blood (Figure 1). The PT and aPTT, measured in-house, were both > 120 sec (reference ranges, 9–12 sec and 59–87 sec, respectively). Due to financial constraints and because frequent plasma transfusions would be needed for future bleeding episodes, the owner elected euthanasia, and a postmortem was declined.

View Figure

• Download as Powerpoint
• Download Figure

Discussion

The author is aware of only one previous case report of afibrinogenemia in the dog.² There are, however, other authors’ personal observations regarding dysfibrinogenemia in a collie and a borzoi.^3,4 One case report of hypofibrinogenemia in a Saint Bernard has been published.⁵ Also, observations of either hypo- or afibrinogenemia in the Bernese mountain dog, Lhasa apso, vizsla, collie, cocker spaniel, and mixed-breed dogs have been published.^6–10

Clinical signs of afibrinogenemia in humans are intermittent. Human patients with afibrinogenemia tend to have less severe bleeding than patients with hemophilia and can go for months at a time remaining asymptomatic.¹¹ However, life-threatening hemorrhages can occur. Patients can have inappropriate bruising or bleeding from the umbilicus or gums, epistaxis, gastrointestinal bleeding, genitourinary bleeding, or bleeding into the abdomen from splenic rupture. Twenty percent of human patients with afibrinogenemia have hemarthroses.¹¹ The most common cause of death is intracranial bleeding.¹¹ Afibrinogenemia can often cause abortion in the first trimester in pregnant patients.¹¹

In veterinary patients, the clinical signs of afibrinogenemia are very similar to clinical signs in humans. Umbilical bleeding, recurrent hemarthrosis, subcutaneous bleeding, and bleeding from the mucus membranes were observed in a family of Saanen dairy goats with afibrinogenemia.¹² Bleeding due to afibrinogenemia has been identified postovariohysterectomy in a bichon frise that presented for intra-abdominal bleeding with concurrent subcutaneous and gastrointestinal bleeding.²

Several different methods are used to measure fibrinogen. The Clauss method is the most commonly used in laboratories as it is accurate, precise, and has the greatest agreement between laboratories when compared with the Blombäck and Blombäck, Ellis and Stransky, chromatin system, PT-derived, and radial immunodiffusion methods.^13,14 In the Clauss method, thrombin is added to a sample and the clotting time is measured. This time is compared against a curve derived from clotting times of known fibrinogen-containing samples. The PT-derived method measures the PT and optical change over the clotting time. This optical change is then compared with a graph plotted with known fibrinogen sample concentrations. The PT-derived method is not recommended for general use in hematology laboratories.¹³ Clottable protein is most often calculated by the Blombäck and Blombäck method, which is performed by adding thrombin in the absence of calcium, washing the clot, dissolving it in urea, measuring the spectrophotometric absorbance at 282 nm, and calculating the fibrinogen concentration using the extinction coefficient. The Blombäck and Blombäck method is unsuitable for routine use in coagulopathy screening, but can be of occasional use in the investigation of congenital fibrinogen defects.¹³ Immunologic assays, of which the enzyme-linked immunosorbent assay (ELISA) is the most accurate and precise, measure the protein concentration, not the functional activity of fibrinogen. Immunologic assays can give false estimates when degraded forms of fibrinogen are present but remain useful in research studies and can help identify congenital fibrinogen defects. Due to its high cost, the ELISA test is limited to research. Maximum clot firmness, measured by rotational thromboelastometry, has a highly linear correlation with both the Clauss and ELISA tests in human studies.¹⁵ Rotational thromboelastometry used as a point-of-care test for liver transplantation in humans had a sensitivity of 83% and specificity of 75% in predicting hypofibrinogenemia.¹⁶

A diagnosis of a fibrinogen disorder is suspected with increases in PT, aPTT, and TCT together with a fibrinogen level below the detection limit for the given assay.^7,9–11 Differential diagnoses include liver failure, autoimmune disease resulting in fibrinogen antibody formation, DIC, anticoagulant rodenticide toxicity, and either sampling or laboratory error.² Acquired hypofibrinogenemia has been described in humans receiving L-asparaginase secondary to decreased hepatic synthesis of fibrinogen and in aplastic anemia patients receiving antithymocyte globulins and corticosteroids.¹¹ Acquired dysfibrinogenemia in humans is most commonly associated with severe liver disease but has also been described with malignancies that produce abnormal fibrinogen, such as hepatomas and renal cell carcinomas.¹¹ The Chihuahua in this report is thought to have acquired afibrinogenemia due to its age, although a congenital form cannot be completely ruled out. In human medicine, afibrinogenemia is most commonly an inherited disorder.

The first step in the diagnosis of a fibrinogen disorder is to rule out more common reasons for prolonged bleeding, such as vitamin K antagonist rodenticide toxicity, DIC, and hemophilia. More common reasons for decreased fibrinogen production, such as liver disease and increased fibrinogen consumption (including DIC and antifibrinogen antibody formation) should also be ruled out. The patient in this case has no history of either ingestion or exposure to medications or toxins prior to the bleeding tendencies described above. The owners changed places of residence and discarded all cleaners when they thought there might have been repeated rodenticide ingestion. All bleeding episodes in this case were incited by either minor trauma or vomiting from dietary indiscretion. The patient’s plasma was mixed with a control patient’s plasma to help rule out fibrinogen autoantibodies. Unlike in the previously reported bichon frise with afibrinogenemia secondary to fibrinogen autoantibodies, the tests performed in this case were not supportive of antibodies directed at fibrinogen.² Further tests, such as a western blot, flow cytometry, or ELISA tests, were not completed and may have shown evidence for fibrinogen autoantibodies.

Serum bile acids and coagulation factors II and VII–XII were within normal limits, ruling out liver failure, rodenticide toxicity, and DIC. Results of D-dimer testing were slightly increased in one blood sample. Given that all other coagulation factors were within normal limits, the slight increase in D-dimer was more likely due to the previous plasma transfusions than from DIC. A control sample was submitted with the patient’s sample to rule out sampling and laboratory errors.

The second step in the diagnosis of a fibrinogen disorder is to differentiate between afibrinogenemia, hypofibrinogenemia, and dysfibrinogenemia. Hypofibrinogenemia is defined as fibrinogen levels between 50 mg/dL and the lower end of normal for the laboratory used.¹⁶ If the fibrinogen levels are < 50 mg/dL, an immunologic test for fibrinogen (such as the ELISA test previously described) is used. Dysfibrinogenemia is diagnosed in patients with lower than normal or normal levels of fibrinogen (measured by an immunologic method) and a functional fibrinogen test (such as the Clauss method) that is below the detection limit.¹¹ A diagnosis of afibrinogenemia is achieved if the results of the immunologic fibrinogen test are < 10 mg/dL and a test measuring the function of fibrinogen is below the detection limit.¹¹ Although those definitions appear to be the ideal means of diagnosing afibrinogenemia, another article states that ruling out liver dysfunction, DIC, and more common coagulopathies with either the Clauss or clottable protein tests below the detection limit and a bleeding history consistent with afibrinogenemia is sufficient for diagnosis of afibrinogenemia.¹⁷ In human medicine, dysfibrinogenemia is unlikely to cause bleeding unless it occurs either postpartum or after a surgical procedure, and most times bleeding is not life threatening.¹¹ The history of severe bleeding, along with the previous Clauss tests results that were below the detection limit, helped with the diagnosis of afibrinogenemia in the dog described in this report.

The author is not aware of any published treatment recommendations specific for fibrinogen replacement in veterinary medicine. Current recommendations in the veterinary literature concerning plasma transfusion rates for coagulopathies range from 15 to 30 mL/kg/day until PT and aPTT are within normal limits.^6,9,10,18 In the case described herein, initial plasma transfusions of 20–30 mL/kg were administered three times and were found to be insufficient for preventing continued bleeding the following day. Three transfusions of ≥ 54 mL/kg were sufficient to prevent further bleeding for up to 2 mo. Similar findings were reported in a bichon frise with afibrinogenemia that presented to a referral hospital postovariohysterectomy.² That dog received a whole blood transfusion (150 mL) on days 5 and 7 of hospitalization, but then required two additional whole blood transfusions (150 mL) over 6 hr on day 9 postsurgically before becoming stable enough to be sent home.

Fibrinogen shows many similarities between dogs and humans. Both species have similar normal reference ranges, and fibrinogen levels < 50 mg/dL are associated with bleeding episodes. Further, similar clinical signs and response to treatment are reported. In human medicine, the end goal for fibrinogen replacement is blood levels of 100 mg/dL with minor bleeding and 200 mg/dL with major bleeding. Therefore, extrapolating the recommended end goals for fibrinogen replacement in human medicine is likely relevant in dogs. Considering the average fibrinogen level in a normal patient is 300 mg/dL, the average plasma volume of a dog is ∼ 50 mL/kg, and the fibrinogen level in a patient with congenital afibrinogenemia is ∼ 0 mg/dL or slightly higher, a 30 mL/kg/day plasma transfusion would yield a calculated 90 mg/dL of fibrinogen, which is an insufficient level to prevent even minor bleeding. Due to recommendations in human medicine, the two case reports in veterinary medicine, and the preceding calculations, the author recommends that afibrinogenemic patients with normal cardiac and renal parameters undergo more aggressive plasma transfusion therapy: 40–60 mL/kg administered over 1–2 days to prevent further bleeding and prolonged hospital stays. A small blood sample should then be submitted for fibrinogen concentration and used as a guideline for future transfusions.

In human medicine, cryoprecipitate is used in place of either fresh or fresh frozen plasma to treat afibrinogenemia to decrease the risk of volume overload. Cryoprecipitate is the treatment of choice in animals as well, but many emergency hospitals do not have cryoprecipitate on hand due to its expense. The recommended cryoprecipitate transfusion volume is 1 U/2.5–5 kg (prepared from 200 mL of plasma). Alternatively, fibrinogen concentrate administrations are used in human medicine in Canada and parts of Europe, and further reduce the likelihood of fluid overload. This product is not available in the United States and is not described as a treatment of dogs. Finally, tranexamic acid^g can be used prior to dental extractions, surgery, or menstruation in patients with low fibrinogen. It acts as an antifibrinolytic by competitively inhibiting the activation of plasminogen to plasmin; however, this increases the risk for thrombus formation. Due to the higher volumes of plasma needed for treatment of afibrinogenemia and the increased risk for thrombus formation, tranexamic acid should be used with care if given concurrently with plasma transfusions.

Conclusion

Afibrinogenemia in dogs is similar to that seen in humans. Further studies are needed, but more aggressive plasma replacement may be needed in dogs with afibrinogenemia than what is recommended with other coagulopathies. The end goal of fibrinogen replacement should not be normalization of PT and aPTT, but instead should be blood levels of fibrinogen between 100 mg/dL and 200 mg/dL, depending on the severity of bleeding. When using plasma, volumes of up to 40–60 mL/kg over 1–2 days will likely be needed for resolution of bleeding. When using cryoprecipitate, the transfusion volume will likely need to be 1 U/2.5–5 kg (prepared from 200 mL of plasma).