Introduction

Aortic thrombosis is an uncommon disease, believed to be associated with either localized or systemic prothrombotic conditions in the distal aorta, where thrombi may form in situ or embolize from the left atrium to the aorta.^1–4 Aortic thromboembolism (ATE) is a critical complication linked to cardiomyopathy in cats and is one of the most common causes of hind limb paresis.⁴ This condition is believed to result from blood stasis in the left auricular appendage.⁴ However, ATE in dogs is often associated with prothrombotic states, including protein-losing nephropathy, protein-losing enteropathy (PLE), prolonged use of steroids, various endocrinopathies, infective endocarditis, and neoplasia. Occasionally, no clear systemic disease can be identified.^1–3,5

Current treatments for ATE include medical management with antiplatelet and anticoagulant medications, together with systemic thrombolytic therapy using tissue plasminogen activator or urokinase infusion.^1,2,5,6 Furthermore, there have been case reports of surgical removal of thrombi in veterinary patients.^7,8 However, evidence supporting the effectiveness of thrombolytic therapy for naturally occurring ATE in dogs remains limited.⁵ For acute limb ischemia in humans, which is similar to acute aortic thrombosis in dogs and cats, endovascular or surgical thrombectomy, catheter-directed thrombolysis, and surgical arterial repair or bypass have become standard approaches.⁹ Although interventional techniques, including rheolytic thrombolysis and vascular stenting, have been reported in veterinary medicine, the optimal interventional therapy for ATE in dogs and cats remains undetermined.^10–12 To the authors’ knowledge, this is the first report of mechanical thrombectomy using a stent-retriever thrombectomy device for ATE in veterinary medicine.

Case Report

A 12 yr old, 6.11 kg, neutered male spitz dog was referred to the animal hospital for acute hind limb paralysis that began 3 hr before admission. The dog had been diagnosed with hyperadrenocorticism and was being treated with trilostane. Two months before the presentation, the patient developed hypoalbuminemia (1.2 g/dL; reference range, 2.3–4.0 g/dL) due to PLE. The patient was being managed for PLE through a prescribed diet and probiotics, without the administration of steroids or immunosuppressants. At presentation, the dog was alert but exhibited severe hind limb pain and tachypnea. Physical examination revealed hind limb paralysis; the limbs were cold to the touch and the foot pads appeared cyanotic. The rectal temperature was low at 36.6°C, and there was a loss of deep pain response in both hind limbs. Lung and cardiac auscultation findings were unremarkable. The initial complete blood count and coagulation tests, including prothrombin time and activated partial thromboplastin time, were within normal ranges. Serum biochemistry revealed a mild increase in alanine aminotransferase (458 U/L; reference range, 19–70 U/L) and total bilirubin (0.7 mg/dL; reference range, 0–0.4 mg/dL) levels and a marked increase in alkaline phosphatase level (1667 U/L; reference range, 15–127 U/L). The serum albumin level at presentation was within the normal range (3.1 g/dL; reference range, 2.6–4 g/dL). The serum lactate was also within normal range, and serum glucose level was mildly elevated (Table 1).

TABLE 1 Venous Electrolytes, Blood Glucose, and Lactate Results at Presentation, 8 Hours After Presentation, and 6 Hours and 24 Hours After the Procedure

View Fullscreen

Abdominal ultrasound revealed large isoechoic to hyperechoic thrombi at the level of the aortic trifurcation, involving both the external and internal iliac arteries (Figure 1). This was accompanied by the absence of vascular signal in both color and pulsed-wave Doppler imaging, indicating no blood flow to the hind limbs. Additionally, thickening of the duodenal and jejunal muscular layers was observed, accompanied by edematous changes in the mesenteric fat and a small volume of anechoic ascites. These findings were consistent with the PLE, diagnosed 2 mo earlier. Increased liver echogenicity with fine texture changes was also noted. A brief echocardiographic examination performed at the referring hospital revealed no abnormalities.

View Figure

• Download as Powerpoint
• Download Figure

The initial treatment included IV administration of fentanyl citrate^a for analgesia, starting with a loading dose of 0.002 mg/kg, followed by a continuous rate infusion (CRI) at 0.002 mg/kg/hr. Additional treatments included oral administration of rivaroxaban^b (2 mg/kg; q 12 hr) and clopidogrel^c (2 mg/kg; q 12 hr) and a subcutaneous injection of dalteparin sodium^d (150 U/kg; q 8 hr). Eight hours after admission, no improvement in the patient’s clinical signs was observed, and lactate levels in both hind limbs had increased from the initial values (right hind limb, 7.5 g/dL; left hind limb, 6.7 g/dL). After obtaining an informed consent from the owners, mechanical thrombectomy using a stent-retriever thrombectomy device was performed. Before anesthesia, premedication was administered IV, consisting of dexamethasone disodium phosphate^e (0.2 mg/kg), as an anti-inflammatory drug, and cefazolin sodium^f (25 mg/kg). For sedation and analgesia, midazolam^g (0.2 mg/kg) was administered IV, along with a CRI of fentanyl citrate. General anesthesia was induced with an IV injection of 1% propofol^h (4.5 mg/kg) and maintained with isoflurane in oxygen. Subsequently, a crystalloid fluid solutionⁱ was administered IV at a rate of 18.3 mL/hr for intraoperative fluid management.

The thrombectomy was performed with the patient placed in dorsal recumbency. A 3 cm incision was made between the trachea and jugular vein on the left side. Subsequently, the subcutaneous tissues were bluntly dissected, and the carotid sheath was carefully separated to expose the carotid artery. After ligating the cranial part of carotid artery, a transarterial approach via the left carotid artery was performed caudal to the ligation suture using a 5-French microvascular access kit^j. A 0.035-inch guidewire^k was advanced through the carotid artery and thoracic aorta and into the abdominal aorta, reaching the level of the third lumbar vertebra (L3) under fluoroscopic guidance^l. The microvascular access sheath was then replaced with a 6-French, 45 cm–long introducer sheath^m, positioning the tip at the L3 level of the aorta. Angiography using a 50% iohexol contrast mediumⁿ was performed to identify the extent of the aortic thrombi and affected vessels (Figure 2). The terminal aorta angiogram revealed a filling defect cranial to the aortic trifurcation, extending distally into both the external and internal iliac arteries. To navigate through the occluded segment, a combination of a 0.016-inch microwire^o and 2.3-French microcatheter^p was used to access both the external and internal iliac arteries. Subsequently, the microwire was withdrawn from the microcatheter, and a 6 mm × 40 mm stent-retriever thrombectomy device^q was advanced through the microcatheter and deployed at the occlusion segments under fluoroscopic guidance. The microcatheter was then retracted, deploying the stent to capture the thrombi, leaving it in place for 60 s. The stent-retriever device was positioned at least 5 mm distal to the end of the occlusion segment in both the external and internal iliac arteries. The device and microcatheter were carefully withdrawn into the sheath, while simultaneous aspiration using a 20 cc syringe was performed through the long sheath sidearm. Suctioning was performed continuously from the moment the stent retriever was retracted into the sheath until it was fully retracted. This process was repeated three times, targeting both the external and internal iliac arteries to ensure thrombi removal. In the first attempt in the right external iliac artery and distal aorta, dark red thromboembolic material was captured within the stent-retriever device. Subsequent attempts captured smaller volumes of thromboembolic material. Following retrieval, the microcatheter was removed, and a repeat angiography through the long sheath sidearm showed improved blood flow and opacification of both the external and internal iliac arteries. The introducer sheath was then removed, the carotid artery was ligated, and the incision in the carotid region was closed using routine surgical techniques. The total procedure time was 61 min, and the anesthesia time was 75 min.

View Figure

• Download as Powerpoint
• Download Figure

After thrombectomy, the dog recovered from anesthesia without any complication. The patient was closely monitored for electrolyte levels after the procedure, and no adverse event such as pain, arrhythmia, or major hemorrhage was observed. To manage postprocedural discomfort and prevent potential complications, IV CRI of fentanyl citrate was administered and gabapentin^r (10 mg/kg; q 12 hr), clopidogrel (1 mg/kg; q 12 hr), rivaroxaban (2 mg/kg; q 24 hr), and amoxicillin-clavulanic acid^s (12.5 mg/kg; q 12 hr) were prescribed. Furthermore, an IV infusion of N-acetylcysteine^t (140 mg/kg; q 12 hr) and hyperbaric oxygen therapy for 45 min every 24 hr were administered. During hospitalization, the dog underwent follow-up examinations, including electrocardiography, complete blood count, venous blood gas analysis, and measurements of blood glucose and lactate levels. Physical examinations immediately after the thrombectomy revealed palpable femoral pulses in the left hind limb, which was warm to the touch. In contrast, the right hind limb remained cold. A repeat abdominal ultrasound, performed 4 hr after the thrombectomy, revealed resolution of the thrombi from the distal aorta to aortic trifurcation, with blood flow signals detected in both the external and internal iliac arteries following thrombi removal. The following morning, palpable femoral pulses and warmth were also noted in the right hind limb, indicating restored circulation. The dog was discharged from the hospital 3 days after the thrombectomy, and the following medications were prescribed: gabapentin (10 mg/kg; q 12 hr), amoxicillin-clavulanic acid (12.5 mg/kg; q 12 hr), clopidogrel (1 mg/kg; q 12 hr), rivaroxaban (2 mg/kg; q 24 hr), and famotidine^u (0.5 mg/kg; q 12 hr). Mild subcutaneous bleeding occurred around the surgical site after the thrombectomy, which was managed with a soft padded bandage placed at the cervical region for 2 days. The affected areas subsequently exhibited ecchymosis and gradually healed.

After discharge, the dog showed resolution of neurological deficits and was able to ambulate with assistance by postoperative day 4 and without assistance by postoperative day 8. Follow-up examinations at 2, 4, and 12 wk, as well as at 6 mo after the procedure, confirmed the maintenance of normal motor function in both hind limbs. The dog continued receiving ongoing medication with trilostane^v (2.2 mg/kg; q 12 hr) and clopidogrel (1 mg/kg; q 12 hr). Moreover, there was no evidence of recurrent ATE in the distal aorta or iliac arteries up to postoperative day 207.

Discussion

The stent-retriever thrombectomy device is a self-expanding stent that can be fully deployed and then retrieved.¹³ In humans, stent-retriever thrombectomy is one of the primary treatment strategies for large-vessel occlusion in acute ischemic stroke.¹⁴ The device used in this study, the Solitaire device, was initially designed and used in Europe to treat patients with ischemic stroke. The first human clinical trials were conducted in 2010.¹³ Meta-analyses involving human patients have demonstrated that successful revascularization was achieved in 77% of patients treated with the Solitaire device. Furthermore, no significant difference in the rates of symptomatic intracerebral hemorrhage or overall mortality was observed between the treatment groups.¹⁵ In another randomized clinical trial, the average number of attempts required for stent-retriever thrombectomy in patients with large-vessel occlusion after acute stroke was 2.67 ± 0.56.¹⁶ In the present case, successful recanalization of the aortic trifurcation and external iliac arteries was achieved after three attempts, which aligns with the average number of attempts typically required for successful thrombus revascularization in humans.

Recommendations for the initial therapy in dogs with ATE are complex owing to the lack of proven therapy.⁵ The clinical differences between acute and chronic presentations have been widely recognized in veterinary medicine.^1,2,5 In ambulatory dogs with a chronic history of ATE, some authors recommend conservative medical management with antiplatelet and anticoagulant medications.¹¹ Conversely, dogs with acute ATE typically present with more severe clinical signs, including pain, neurologic deficits, and nonambulatory status, and they are less likely to survive to hospital discharge.² A previous retrospective study demonstrated a median survival time of 9 days for dogs with acute ATE compared with 293 days for those with chronic ATE.¹ In this case, the dog presented with acute thrombosis, characterized by acute hind limb paralysis and severe pain upon admission. The thrombus removal through mechanical thrombectomy using a stent-retriever thrombectomy device during the acute phase resulted in the recovery of motor function in both hind limbs by postoperative day 4.

The prognosis for ATE is associated with limited survival in acute and nonambulatory cases, regardless of treatment.^1–3 According to a recent retrospective study of dogs with ATE treated with various therapies, including anticoagulant and antiplatelet agents administration, balloon angioplasty, and thrombolytic therapy, 63% of hospitalized dogs (n = 41/65) survived to hospital discharge. However, only 31.7% (n = 13/41) of those discharged remained alive at 180 days after discharge.² Rheolytic thrombectomy and vascular stenting at the aortic trifurcation have been described as interventional treatments in dogs and cats with ATE.^10,12 In a retrospective study investigating the application of rheolytic thrombectomy in feline ATE, successful thrombus removal was achieved in 83% (n = 5/6) of cats, of whom three were discharged.¹⁰ Another recent study using vascular stenting in seven dogs with ATE reported successful revascularization in all of them, with a median survival time of 264 days. Five of the seven dogs were ambulatory within 2 days of vascular stenting and survived to be discharged.¹² In the current case, vascular stenting was initially considered; however, this option was deemed unfeasible because of the small patient size and the narrow diameter of both external iliac arteries (1.7 mm). However, although very small-diameter stents designed for human coronary arteries were available, their high cost posed a significant challenge. In contrast, the stent-retriever thrombectomy device, which can be deployed through a microcatheter, offered the advantage of being applicable even in small vessels. This adaptability made it a feasible option for this case, contributing to the successful outcome.

Ischemia-reperfusion injury (IRI) is a major concern associated with thrombolytic therapy, often leading to severe complications, including sudden death.⁶ This condition occurs when the recanalization of ischemic or necrotic tissue facilitates the systemic release of potassium and metabolic organic acids. The occurrence of severe IRI following thrombolysis and thrombectomy procedures is a significant complication, documented in numerous cases of feline ATE.^6,10 In the present case, no signs of significant IRI, including hyperkalemia, were observed after the procedure. Treatments with N-acetylcysteine, dexamethasone sodium phosphate, and hyperbaric oxygen therapy were used for their antioxidative and anti-inflammatory effects.¹⁷ However, the clinical efficacy of these treatments in preventing IRI remains uncertain. Moreover, because of concerns about potential periprocedural IRI, we closely monitored serum electrolytes, glucose, and lactate levels and conducted electrocardiographic evaluations. The onset of clinical symptoms in canine ATE is highly variable, with previous studies reporting acute symptoms in 15–47% of dogs and chronic symptoms in 43–69% of them.^1,3,5 Some dogs exhibit chronic signs that cause acute exacerbations.⁵ It is possible that a thrombus formed in situ in the aorta over time, with acute clinical symptoms emerging once blood flow was completely obstructed. In this case, although the clinical symptoms appeared acutely, it remains unclear whether the thrombosis obstructed the aorta after forming elsewhere or if it gradually developed in the aorta, leading to the abrupt onset of symptoms once the vessel was fully blocked. The chronicity of clinical signs suggests a less complete obstruction of blood flow to the pelvic limbs compared with that in dogs with more acute symptoms.⁵

In human neurovascular interventions using stent-retriever thrombectomy devices, recognized complications include embolic migration, intracranial hemorrhage, and vessel wall injuries.^14,15 Vessel perforation during stent-retriever thrombectomy, although rare, may occur, potentially leading to high mortality.¹⁸ A study evaluating vascular damage after stent-retriever thrombectomy in canine stroke models through histopathological analysis revealed no inflammation, hemorrhage, or device-induced medial injury in all cases.¹⁹ In a previous canine stroke model study, only one dog exhibited severe intimal proliferation, marked diffuse thrombosis, and mild endothelial cell loss after 1 mo.¹⁹ Another study comparing vessel wall damage between stent-retriever and catheter aspiration thrombectomy in a canine stroke model found that the stent-retriever group had a significantly increased risk of endothelium denudation compared with the catheter aspiration group, although minimal arterial wall damage was observed in both groups.¹⁴ In the present case, although a histological examination was not performed, no serious vascular damage or complications, including postprocedural thrombosis, vessel stenosis, or perforation, were observed during the procedure.

The carotid artery cutdown technique for arterial catheterization was used to access the distal aorta for thrombectomy in this patient. Ligation of the carotid artery has been experimentally demonstrated to be tolerable in dogs.²⁰ In this case, no major complications occurred at the access site, despite the patient being on prior antiplatelet and anticoagulant medications. Mild subcutaneous bleeding was observed, which was managed with a soft padded bandage for 2 days. In dogs, the options for closing vascular access sites include arterial ligation, vascular repair, manual compression, or the use of vascular closure devices. However, because of their high cost, vascular closure devices are rarely used in veterinary clinics and are generally not recommended for closing the carotid artery. In the present case, surgical ligation was performed to minimize bleeding complications, and no major complications were observed at the surgical site, apart from mild subcutaneous bleeding.

Conclusion

This case report suggests that mechanical thrombectomy using a stent-retriever thrombectomy device is a viable treatment option for dogs with acute-stage ATE, offering an alternative to conventional therapeutic approaches. Future prospective studies are necessary to assess the mortality and complication rates associated with the use of a stent-retriever thrombectomy device for mechanical thrombectomy in veterinary patients.

The authors would like to thank Songhui Lee, in the Department of Clinical Sciences, College of Veterinary Medicine and Research Institute for Veterinary Science, Seoul National University, for assistance with illustrations.

Written consent from the dog’s owner was obtained for medical data to be used for educational and research publication purposes. This study is a case report of canine patients, and identification information, such as patient name, was excluded. All images used in the manuscript are images created by the author during the diagnosis and treatment process.