Use of the D-dimer Assay for Diagnosing Thromboembolic Disease in the Dog

Introduction

Evidence and Strategies of Noninvasive Testing

The history and clinical presentation may raise the suspicion for PTE, but these findings in animals are inconsistent and nonspecific. Likewise, routine laboratory tests, electrocardiography, thoracic radiography, and blood gas analyses are not reliable tests to confirm or rule out PTE. The presence of identifiable risk factors may be the clinician’s best indication to screen for underlying PTE complications. Risk factors for people with embolic complications are well documented. Table 1 lists the identified factors associated with a high risk of developing thromboembolism in humans compared to dogs.^{1,4,6,7,11,12} The list of risk factors for dogs is not quite as substantial as that for humans, possibly because a confirmed diagnosis of PTE is not made in a significant number of cases. A recent investigation revealed that 60% of thromboembolism cases diagnosed in dogs had underlying protein-losing nephropathy or neoplasia.¹³ This finding underscores the need for clinicians to suspect embolic processes, particularly in animals with glomerular disease and cancer.

In humans, advanced imaging studies have caused a major improvement in the diagnosis of embolism. Ventilation-perfusion scanning has played a central role and has proven to be a valuable tool when the results are definitive.¹⁴ A normal ventilation-perfusion scan essentially rules out the diagnosis of PTE; however, large studies have disclosed that most humans with suspected embolism have ventilation-perfusion scans that are not definitive.¹⁰ For example, many patients with primary respiratory diseases who must have their symptoms differentiated from those caused by PTE, have some degree of respiratory pathology that contributes to confusing or inconclusive ventilation-perfusion scan results. Ventilation-perfusion scans are often scored as having low, intermediate, or high probability of PTE, and this score is factored into the results of other tests to create a picture that is supportive of or rules out PTE.^7,10,14

The use of helical computed tomography (CT) has also been a major advance in the diagnosis of embolism in humans. In contrast to ventilation-perfusion scanning, it allows direct visualization of the thrombi. Studies have shown the sensitivity and specificity of helical CT scanning ranges between 70% and 100%, and newer, faster scanners are now providing even greater resolution.^15,16 Unfortunately, a normal helical CT scan cannot be used to rule out PTE with the same degree of certainty that a negative ventilation-perfusion scan provides.^17,18 Clinical trials describing the use of ventilation-perfusion or helical CT scans to detect PTE in animals have not been performed to date.

In clinical veterinary practice, echocardiography is often utilized to help identify supportive evidence for PTE. Evidence of pulmonary arterial hypertension derived from a thorough echocardiogram and Doppler study may support the diagnosis of PTE. Echocardiographic findings of pulmonary arterial hypertension may include a poorly contracting right ventricle, leftward shift of the septum during part of the cardiac cycle, pulmonary artery dilatation, and high-velocity tricuspid and pulmonic regurgitation jets. The regurgitant jets can be used to estimate right heart pressures, and in the absence of pulmonic stenosis, they indicate pulmonary arterial hypertension.¹⁹ Although the identification of pulmonary arterial hypertension is not analogous to the diagnosis of PTE, it may increase the probability of embolism when combined with other diagnostic methods. In animals with significant pulmonary hypertension and no obvious respiratory pathology on thoracic radiographs, PTE is considered a likely differential diagnosis.

D-dimer Assay

Laboratory markers of coagulation, such as fibrinopeptides A and B, fibrinogen degradation products (FDPs), prothrombin fragments 1 and 2, and thrombin-antithrombin complexes, have been intensely investigated for prediction of thromboembolic disease.^9,11,12 Of the available laboratory markers, only D-dimer has shown clinical utility in detecting early embolism in humans.^9,20 Fibrin formation (i.e., the result of thrombin formation on circulating fibrinogen) stabilizes the clot structure. The initial fibrin monomers are cross-linked by the action of thrombin-activated factor XIII. Plasmin indiscriminately degrades fibrinogen and cross-linked fibrin monomers in stable clots. One of the FDPs is fragment D. D-dimers result only from the degradation of cross-linked fibrin (i.e., stabilized clot), and in contrast to other degradation products, they are specific for active coagulation and fibrinolysis.^21–23

A few studies have documented the utility of D-dimer concentrations in small animals. One investigation evaluated the sensitivity and specificity of FDPs and D-dimer concentrations in 20 dogs with disseminated intravascular coagulopathy (DIC).⁴ This study concluded that the latex agglutination D-dimer test could replace the FDP test for diagnosing fulminant DIC.⁴ The D-dimer assay was more sensitive than FDP assays and equally specific.⁴ Other preliminary veterinary studies have shown promise for using D-dimer to screen for DIC and thromboembolic disease prior to overt DIC.^3,5,13,24 A recent prospective study designed to investigate D-dimer concentrations in clinically ill dogs, sought to determine if the marker could differentiate dogs with pathological thromboembolism from dogs with other clinical problems (e.g., cardiac, hepatic, renal, and neoplastic diseases) and from dogs postoperatively.¹³ Although there was considerable overlap in D-dimer concentrations between illness groups, D-dimer concentrations were highest in dogs with embolic disease, followed by dogs in the hepatic disease group.¹³ Only these two groups had median D-dimer concentrations significantly different from clinically healthy dogs [Table 2].¹³ The neoplastic disease group demonstrated only a marginal difference from the control group.¹³ The overlap between D-dimer concentrations was not surprising, because both hepatic disease and neoplasia may be associated with thromboembolism, coagulopathies, or hemorrhage from vascular damage or destruction. Two dogs in the neoplastic disease group and one dog in the hepatic disease group presented with hemoabdomen. These dogs had some of the highest D-dimer concentrations in their respective group.¹³ High D-dimer concentrations are expected in dogs with body cavity hemorrhage and may not be indicative of pathological thromboembolism.^13,24

In this prospective investigation, the sensitivity of D-dimer concentrations >500 ng/dL for predicting thromboembolic disease was 100%. The specificity of D-dimer for thromboembolism at that concentration was 70%.¹³ The specificity of D-dimer concentrations >1000 ng/dL in predicting thromboembolism was 94% (sensitivity 80%), and the specificity of concentrations >2000 ng/dL was 98.5% (sensitivity 36%).¹³ A latex agglutination assay was utilized for this study, but in humans the enzyme-linked immunosorbent assay (ELISA) for D-dimer has been shown to be more sensitive.²⁵ The decreased specificity of D-dimer concentrations for thromboembolic disease, particularly at the lower levels, has been previously observed in humans.^20,21 When selecting a screening test for pathological thrombi, the sensitivity of the test is probably more important than the specificity. While false positives are undesirable, false negatives are potentially fatal. Fibrin degradation products were not elevated in any dog with pathological thrombi in the study described above, and FDPs may be insensitive indicators of thromboembolism, with or without overt DIC. No dog with thromboembolism had a negative D-dimer concentration in the study, suggesting that false negatives are uncommon.¹³ A negative D-dimer concentration may, in effect, exclude thromboembolism in a suspected case. This strategy for interpreting D-dimer results is well accepted in humans.^22,25–27

Diagnostic Approaches for Thromboembolic Disease

The initiation of diagnostic tests for PTE relies on a clinical suspicion of the disease. Animals with known risk factors for embolism, particularly glomerular disease and cancer, should increase the clinician’s degree of suspicion. Given that no test is 100% sensitive and specific for thromboembolic disease, the author recommends adopting the approach of building a “clinical probability picture” based upon scoring the animal as having a low, intermediate, or high clinical probability of thromboembolic disease. The algorithm depicted in the Figure demonstrates the suggested approach to an animal with suspected thromboembolic disease. This approach is simple and includes the use of clinical and diagnostic equipment available in most clinical veterinary practices. Practices equipped with advanced imaging technology as described earlier may also be able to acquire additional evidence for the exclusion or diagnosis of thromboembolism.

Conclusion

Although the exact incidence of PTE in animals is unknown, it is thought that thromboembolism is a substantial, underdiagnosed complication of seriously ill animals. When the diagnosis of embolism is made, it is often late in the course of the disease process, and the prognosis is poor. Because early PTE may exhibit vague or subtle clinical signs, the presence of identifiable risk factors may be the clinician’s best indication to screen for underlying thromboembolic complications. A sensitive D-dimer assay may essentially rule out thromboembolism if negative. If positive, it may help the veterinarian assess the likelihood of thromboembolic complications and guide the next diagnostic and therapeutic steps. The goal of the veterinarian is to recognize the critically ill animal with DIC or thromboembolic disease that would benefit from early anticoagulant therapy, thus reducing the significant morbidity and mortality associated with this severe complication.