Outcome and treatments of dogs with aortic thrombosis: 100 cases (1997‐2014)

Abstract Background Aortic thrombosis (ATh) is an uncommon condition in dogs, with limited understanding of risks factors, outcomes, and treatments. Objectives/Hypothesis To describe potential risk factors, outcome, and treatments in dogs with ATh. Animals Client‐owned dogs with a diagnosis of ATh based on ultrasonographic or gross necropsy examination. Method Multicentric retrospective study from 2 academic institutions. Results One hundred dogs were identified. Anti‐thrombin diagnosis, 35/100 dogs were nonambulatory. The dogs were classified as acute (n = 27), chronic (n = 72), or unknown (n = 1). Fifty‐four dogs had at least one comorbidity thought to predispose to ATh, and 23 others had multiple comorbidities. The remaining 23 dogs with no obvious comorbidities were classified as cryptogenic. Concurrent illnesses potentially related to the development of ATh included protein‐losing nephropathy (PLN) (n = 32), neoplasia (n = 22), exogenous corticosteroid administration (n = 16), endocrine disease (n = 13), and infection (n = 9). Dogs with PLN had lower antithrombin activity than those without PLN (64% and 82%, respectively) (P = .04). Sixty‐five dogs were hospitalized with 41 subsequently discharged. Sixteen were treated as outpatient and 19 euthanized at admission. In‐hospital treatments varied, but included thrombolytics (n = 12), alone or in combination with thrombectomy (n = 9). Fifty‐seven dogs survived to discharge. Sixteen were alive at 180 days. Using regression analysis, ambulation status at the time of presentation was significantly correlated with survival‐to‐discharge (P < .001). Conclusions/Clinical Importance Dogs with ATh have a poor prognosis, with nonambulatory dogs at the time of presentation having worse outcome. Although the presence of comorbid conditions associated with hypercoagulability is common, an underlying cause for ATh was not always identified.

K E Y W O R D S canine, protein-losing nephropathy, saddle thrombus, thromboembolism, thrombus

| INTRODUCTION
Aortic thrombosis (ATh) is an uncommon condition in dogs, thought to be associated with a regional prothrombotic environment in the distal aorta. [1][2][3] This is in contrast to aortic thromboembolism (ATE) in cats, where thrombi formed in the left atria or auricle embolize to the aorta. 4 Also in contrast with ATE in cats, overwhelmingly because of an underlying cardiomyopathy 4,5 , a variety of diseases have been implicated in ATh in dogs, including protein-losing diseases, endocrinopathies, and neoplasia. [1][2][3][6][7][8][9][10][11][12][13][14] In a previous study, 58% of dogs with ATh have no identified underlying disease, 1 suggesting that ATh with no identifiable cause (or cryptogenic) might be common. Despite an overall guarded prognosis for dogs with ATh, long-term survival occurs in some dogs, suggesting heterogeneity in the clinical presentation of affected dogs. 2 Consensus guidelines on the optimal strategy for the prevention and treatment of thrombosis in animals have been established, but never studied prospectively. 15 In previous studies, treatment strategies have been clinician-dependent and the variations in cases and treatments make comparison in outcomes difficult. Commonly reported treatment for ATh in dogs typically involves combinations of antiplatelet (eg, clopidogrel, aspirin) alone or in combination with anticoagulant therapies such as heparin, warfarin, direct oral anticoagulants such as rivaroxaban and apixaban [16][17][18] with thrombolysis by tissue plasminogen activator (TPA) being attempted with varying clinical outcomes. [19][20][21][22] The efficacy of these therapies in dogs with ATh has been minimally documented. 1,3 Available literature on ATh in dogs is limited to case reports, descriptive studies on small numbers of dogs, and a few larger studies without information on outcome or treatments. [1][2][3][6][7][8][9][10][11][12][13][14][23][24][25][26][27][28] The aim of our study was to describe the presenting clinical characteristics of dogs with ATh, specifically underlying causes and descriptions of clinical management.

| MATERIALS AND METHODS
The electronic medical records at Cummings School of Veterinary

| RESULTS
From the medical records at TCSVM and OSU-CVM, a total of 100 dogs with ATh were included (61 and 39 dogs, respectively). Aortic thrombosis was diagnosed by ultrasonography in 91 cases and by direct visualization at necropsy in 9 cases. Dogs diagnosed with ATh during necropsy did not have an ultrasound performed but exhibited clinical signs compatible with ATh.

| Onset of clinical signs
The median duration of clinical signs before presentation was 10.5 days (0-912). Twenty-seven dogs were presented with acute signs, 72 with chronic signs, and 1 dog with an unknown duration of clinical signs.
Dogs with acute signs were more likely to be nonambulatory (P < .001), have pelvic limb neurologic deficits (P = .007), have bilateral signs (P = .01), and be painful (P = .03) compared to dogs with chronic ATh (Table 1). Dogs with an acute onset of signs were significantly less likely to survive to discharge (P = .04). However, the percentage of euthanasia-at-admission was similar for dogs with acute signs compared to dogs with chronic signs (22% and 18%, respectively; P = .41).

| Possible causes of ATh
The most commonly diagnosed systemic condition known to be associated with hypercoagulability in our study population was PLN (n = 32), either alone (n = 22) or associated with 1 (n = 8) or 2 other conditions (n = 2; Table 3

| In-hospital treatment
Sixty-five dogs were hospitalized for in-hospital management ( Figure 1). The median duration of hospitalization was 2 days (0-12).

| Outcome
Aspirin + UH (n = 1), clopidogrel + LMWH (n = 2), Riva + LMWH (n = 1) Aspirin + UH (n = 1), aspirin + LMWH (n = 2), aspirin + Warfarin (n = 1), aspirin + warfarin + UH (n = 1), clopidogrel + UH (n = 3), clopidogrel + LMWH (n = 1) Multiple anticoagulant UH + LMWH (n = 1) rivaroxaban + LMWH (n = 1) UH + LMWH (n = 1) No treatment n = 14 n = 3 n = 11 Note: This table does not include the 9 dogs that were lost to follow-up at 180 days. Abbreviations: LMWH, low-molecular weight heparin; UH, unfractionated heparin.  [43][44][45][46] and has been identified as a possible risk factor for ATh in dogs. [1][2][3]7 Although previous studies have proposed a possible association between heart disease and the development of ATh in dogs, 3,6,7 none of the dogs in our study were diagnosed with a structural heart disease thought to be the cause of the ATh, consistent with another recent study where structural heart disease was not identified in any of the 26 ATh dogs studied. 1 However, 4 of our dogs had evidence of infective endocarditis on echocardiogram and 3 had a cardiac mass, so it is possible that some types of cardiac disease might contribute to the development of ATh in dogs. However, in contrast to ATE in cats, that role appears to be minimal. Approximately two-thirds of dogs in our study were treated for ATh, and the treatments were variable, reflecting individual clinicians' preferences. The optimal treatment for ATh is unknown. In ambulatory dogs with a chronic history of ATh, some authors advise medical management using a combination of antiplatelet and anticoagulant medications alongside physical therapy and supportive care. 29 For acutely affected dogs that are severely affected, interventional techniques have been recommended and are described elsewhere. 29 48 Platelet inhibitor drugs were the most common treatment in our study. The rationale for the use of platelet inhibitors is that arterial thrombosis occurs in a high-shear stress environment that promotes platelet hyper-reactivity. 49 Historically, aspirin has been used to minimize the development of recurrence of thromboembolism in ATE in cats.
This treatment has fallen out of favor because clopidogrel is a more effective medication to inhibit platelets, and to delay or prevent a relapse of ATE when compared to aspirin. 16,50 Clopidogrel is also recommended over aspirin for people with acute ischemic stroke. 51 As clopidogrel is the current antiplatelet drug of choice for ATE in cats and has been suggested to be more effective than aspirin in ATh in dogs, use of clopidogrel to treat ATh in dogs seems a reasonable choice. 47  Currently, oral anticoagulants (ie, rivaroxaban, apixiban) have become more common in veterinary medicine because of their ease of administration, predictable pharmacokinetics, and superior effects in treating deep vein thrombosis in humans. 29,54 Thrombolysis was attempted in 12 dogs. A previous study reported 4 dogs with ATh who received systemic TPA, 3 of whom were either died or euthanized due to failure to improve or progression of disease. 1 One case report describes successful decrease in the size of a catheter-associated jugular vein thrombus using TPA. 24 Although the use of TPA and streptokinase in ATE cats has been controversial due to concern for reperfusion injury, 18 In people with an acute ischemic stroke, thrombectomy is typically not recommended due to an increased risk of death in the face of a similar rate of good clinical outcome when compared to medical management. 51 Local delivery via catheter directed thrombolysis is used in humans with severe cases of DVT 56 and has been described in 1 dog with ATh using TPA. 55 It is currently unknown if these same guidelines will be translatable to small animals.
Our study showed a discharge rate of 57%, with 28% of discharged dogs alive at 180 days. Our study identified ambulatory status as the only predictor of outcome. This finding might be related to owners being willing to pursue treatment on less affected dogs, and therefore might not be related to animal characteristics or treatment instituted.
Our study did not identify chronicity as a risk factor for death, contrary to a previous study that documented a difference in median survival times between acute and chronic (ie, clinical signs >24 hours) dogs (9 days vs 293 days, respectively). 2 This difference might be explained by difference in severity of disease, or long-term impact of therapy instituted. Our survival rate is difficult to compare to other studies, as identification of time of death (ie, hospital discharge or final outcome) is lacking in the 2 major retrospective studies. 1,2 Our study has several limitations. As a retrospective study, dogs were not randomized to treatment, and the treatments used varied greatly in the medications, dosing, and outpatient protocols. Some data points were missing, including more thorough TEG evaluation, and some information such as adverse effects of treatment or clot res-

ACKNOWLEDGMENT
Part of these data has been presented at the International Veterinary Emergency and Critical Care Society (IVECCS) meeting, September