Chylothorax, characterized by the accumulation of chyle in the pleural space, is an infrequently reported condition in dogs (5,7,18,19,32,36). It has been suggested that obstruction of either the thoracic duct or cranial vena cava (CrVC) results in stasis of flow within the thoracic lymphatics, leading to dilation, and chyle then leaks out of the dilated lymphatics (4). Reported causes of chylothorax in dogs and cats include neoplasia (14,32), heart disease (3,7,17), lung lobe torsion (18), CrVC thrombosis (5), diaphragmatic hernia (25), fungal infection (19), and congenital anomaly of the thoracic duct (36). However, in most dogs and cats, so far, no obvious cause has been identified, leading to their classification as idiopathic chylothorax (15,16).

CrVC thrombosis is an uncommon and potentially life- threatening complication that is associated with various critical illnesses (31). Thoracic radiography may reveal pleural effusion but lacks specificity in diagnosing CrVC thrombosis (31). Previously reported cases of CrVC thrombosis were mostly diagnosed through ultrasound examinations (5,31,33). Ultrasonographic features of CrVC thrombosis include incomplete or nearly complete obstruction of blood flow (5,27). However, ultrasound examinations have limitations as they cannot differentiate between intraluminal thrombus and intraluminal neoplasia (31). Also, due to the limited capability of ultrasound examinations to inspect only a part of the CrVC, only advanced thrombosis can be detected (27). Computed tomography angiography (CTA) can be used to confirm thrombus within the CrVC, but only a few cases have been diagnosed (11,30).

The purpose of this report is to present a case diagnosed with chylothorax associated with an unusual occurrence CrVC thrombosis in a dog using CTA.

A 13-year-old, castrated male Miniature Pinscher dog, weighing 3.25 kg, presented with coughing and dyspnea. The dog had been coughing for the past 4 weeks. 2 weeks before presentation, pleural effusion was confirmed through thoracic radiographs, and an ultrasound-guided thoracocentesis was performed at a local hospital. Pleural fluid had a characteristic milky appearance, with a high concentration of triglycerides and a low cholesterol:triglyceride ratio. The pleural fluid was characterized as chylous effusion through cytologic, physical, and biochemical evaluations. The patient was referred to the Gyeongsang National University Animal Medical Center for the evaluation of the cause of chylothorax.

The patient had mild dehydration on physical examination, and muffled heart sounds were detected. Normal findings were observed on complete blood count and serum chemistry except for mild thrombocytopenia (164,000 cells/µL, reference range: 200,000-500,000 cells/µL). Thoracic radiography revealed rounded and retracted lung lobes, obliteration of cardiac silhouette and diaphragm, and increased opacity in the thoracic cavity, consistent with pleural effusion. Except for mild hepatomegaly, no abnormalities were observed on the abdominal radiographs. An echocardiogram conducted to differentiate chylothorax caused by heart disease revealed no specific findings other than mild myxomatous mitral valve degeneration.

Computed tomography (CT) lymphangiography was performed using a 160-multislice CT scanner (Aquilion Lightning 160, Canon Medical Systems, Otawara, Japan) to differentiate the underlying cause of chylothorax. General anesthesia was induced using an alfaxalone (2 mg/kg, IV, Alfaxan^®, Careside, Gyeonggi-Do, Korea) and maintained with isoflurane (Irfan^®, Hana Pharm, Kyonggi-Do, Korea) in oxygen (2.0 L/min). CT lymphangiography was performed with intrametatarsal pad injection using 1 mL/kg of nonionic iodine contrast medium (350 mgI/mL, Omnipaque^®, GE Healthcare, Oslo, Norway) with a 22-gauge needle. Subsequently, the metatarsal pads, legs, hocks, and thighs were massaged for 3-4 minutes (24). A CT scan was performed 5 minutes after the intrametatarsal pad injection, and a satisfactory enhancement of the thoracic duct was obtained. Mild dilation of the thoracic duct and lymphangiectasia at the level of the cranial mediastinum were identified, but there was no evidence of obvious leakage of contrast medium indicative of thoracic duct rupture (Fig. 1).

Figure 1. Computed tomography (CT) lymphangiography image of a patient with chylothorax. 5 minutes after injection, a satisfactory contrast-enhanced thoracic duct was visualized. (A) Thoracic duct (arrow) in the transverse CT images at the level of T8. Pleural effusion is evident bilaterally in the ventral portion of the thorax (asterisk). (B) Transverse CT images at the level of the cranial mediastinum reveals lymphangiectasia characterized by dilated and tortuous lymphatic vessels (arrowhead). (C) Sagittal multiplanar reconstruction (MPR) image showing mild dilation of the thoracic duct (arrows) and lymphangiectasia (arrowhead). There is no evidence of a thoracic duct rupture or congenital anomaly as the possible cause of chylothorax.

After CT lymphangiography, CTA was performed in a helical mode with intravenous (IV) contrast administration. A dual injector (Salient^®, Medrad Inc., Pittsburgh, PA, USA) was used to inject 3 mL/kg of nonionic iodine contrast medium (300 mgI/mL, Omnipaque^®, GE Healthcare, Oslo, Norway) and saline solution. CTA revealed an intraluminal filling defect of approximately 3 cm in length within the CrVC, indicating the presence of a thrombus (Fig. 2A, B). There was a small mineral attenuating material in the periphery of the lung parenchyma, which was considered dystrophic mineralization caused by chronic chylothorax. Complete atelectasis was found in the right cranial and middle lung lobes, as well as in the cranial and caudal parts of the left cranial lung lobe, caused by moderate amounts of pleural effusion. Based on radiography, pleural fluid analysis, and CTA, the dog was tentatively diagnosed with a chylothorax associated with a CrVC thrombosis.

Figure 2. Contrast-enhanced computed tomography (CT) multiplanar reconstruction (MPR) images of the transverse and sagittal plane for diagnosis (A, B) and after treatment for thrombosis (C, D). (A, B) Contrast-enhanced CT MPR images with anesthesia for diagnosis reveal a large filling defect (arrow) in the lumen of the cranial vena cava instead of contrast media at the level of T3. Pleural effusion surrounded the cranioventral aspect of the lung lobes (asterisk). (C, D) Contrast-enhanced CT MPR images without anesthesia after treatment for thrombosis showing a reduction in thrombus (arrowhead) size.

On the day of diagnosis, the dog was administered clopidogrel (10 mg/kg, Lopirel Tab, CMG Pharma, Gyeonggi-Do, Korea) orally. Starting the following day, the dog was administered clopidogrel (2 mg/kg, PO, q 24 h) and rivaroxaban (1 mg/kg, PO, q 12 h, Xarelto^®, Bayer AG, Berlin, Germany). Despite the thrombosis treatment, the amount of pleural effusion did not reduce, prompting plans for a pleural port placement. One week before the placement of the pleural port, the administration of clopidogrel and rivaroxaban was discontinued, and subsequently, edema was observed in the neck, chin, and forelimbs. After the placement, clopidogrel and rivaroxaban were continued at the same dosage, and although the edema was reduced, it was still observed. The amount of pleural effusion increased steadily, reaching up to 350 mL per day. Consequently, thrombolytic treatment using a recombinant tissue-plasminogen activator (rt-PA, Actilyse, Boehringer Ingelheim, Ingelheim, Germany) was attempted 1 month after the diagnosis. One dose of rt-PA (0.2 mg/kg) was administered at once, followed by additional doses of 0.4 mg/kg every 60 minutes for a total of four administrations. A second course of rt-PA was administered at a dose of 1 mg/kg every 60 minutes for a total of four administrations. There was no hemorrhage at the entry sites of the catheter and pleural port. The next day, an ultrasound examination revealed a decrease in the size of the thrombus within the CrVC, and the edema observed in the neck, chin, and forelimbs was resolved.

The amount of pleural effusion diminished steadily during the 2 months following rt-PA administration, and 2 months after treatment, fluid production was 9 mL per week. Three months post-treatment, a follow-up CTA was performed without anesthesia to assess the patient’s condition. CTA examination revealed that the length of the thrombus, initially measured at 3.1 cm, had decreased to approximately 1.3 cm at the same location (Fig. 2C, D). Despite the presence of a small residual thrombus in the CrVC and the ongoing minimal pleural effusion, the patient survived for 18 months after the diagnosis. However, the patient later died due to a trauma unrelated to thrombosis.

CrVC thrombosis has been reported as one of the causes of chylothorax in dogs (31). Concurrent illnesses, including immune-mediated disease, sepsis, neoplasia, fungal infection, protein-losing nephropathy and hyperadrenocortism varied among dogs with thrombosis (19,31,32,35,37). Additionally, most of the cases reported previously confirmed a history of central IV catheter placement (31). In a study of 17 dogs with CrVC thrombosis, it was found that in all cases, there was a history of central IV catheter placement before thrombus formation, including the use of a 19-gauge through-the-needle catheter, cardiac catheter, and subcutaneous vascular access port (31). This implies that central IV catheter placement could be a risk factor for thrombus formation (31). Therefore, central IV catheter placement should be approached with caution in patients at high risk of thrombosis (31). In this case, the patient had no history of central IV catheter placement, and there were no confirmed underlying diseases, such as immune-mediated disease, cardiac disease, neoplasia, infectious diseases, sepsis, protein-losing nephropathy, or hyperadrenocortism, that could induce thrombosis. Therefore, the cause of thrombosis could not be determined.

CrVC syndrome is characterized by a combination of clinical signs arising from partial or complete obstruction of the CrVC (29). When external compression, invasion, or intraluminal obstruction occurs in the CrVC, venous flow from the cranial portion of the body is impaired (29). Typical clinical signs of CrVC syndrome include subcutaneous pitting edema of the head, neck, and forelimbs, with or without pleural effusion (29). In this case, it is likely that the thrombus within the CrVC impaired venous return, resulting in edema in the neck, chin, and forelimbs, consistent with CrVC syndrome. Also, it was considered that the resolution of CrVC syndrome occurred as the size of the thrombus decreased after thrombolytic treatment.

Anatomy of the canine thoracic duct and its branches is highly variable among dogs, and lymphangiography is used to evaluate the anatomical features of the thoracic duct (22). Lymphangiography differentiates the rupture or congenital anomaly of the thoracic duct, which is the cause of chylothorax (2,36). Also, preoperative lymphangiography is recommended for appropriate surgical planning (8,13). CT lymphangiography can more accurately identify branches of the thoracic duct than the standard radiographic lymphangiography (13,34). To perform lymphangiography, various contrast medium injection sites have been reported, including the mesenteric lymph node (6,21), popliteal lymph node (10,28), perianal subcutaneous region (1,20), dorsal metatarsal region (23), and intrametatarsal pad (9,24). In this case, the patient was a small dog weighing 3.25 kg, making it challenging to inject a large amount of contrast medium into the lymph nodes. Therefore, the contrast medium was injected through the intrametatarsal pad, which is a feasible and relatively easy procedure with a reported high success rate (24). The thoracic duct was successfully opacified when 1 mL/kg of the contrast medium was injected and massaged, and scanning was performed 5 minutes after the contrast injection. In a previous study that performed CT lymphangiography in dogs with chylothorax, findings of lymphangiectasia or a ruptured thoracic duct were observed at the level of the cranial mediastinum (2,23). In this case, mild dilation of the overall thoracic duct was observed during the CT lymphangiography examination. This is considered to be a result of increased pressure within the CrVC due to a thrombus, leading to obstruction in the outflow of the thoracic duct.

The majority of CrVC thrombosis cases are diagnosed through ultrasound examinations such as thoracic ultrasonography, echocardiography, and transesophageal echocardiography (5,26,31,33). However, given the limited visibility of the CrVC on ultrasound examinations, only more advanced thrombosis may be detectable. In addition, ultrasound examinations are not capable of distinguishing between an intraluminal thrombus and intraluminal neoplasia (27,31). CTA can complement these limitations and assess potential locations of thrombus formation. Cases confirming CrVC thrombosis through CTA are rarely reported in dogs. Among them, one case confirmed a thrombus through ultrasonography, and a CTA scan was conducted for precise evaluation (11). In another case, a CT examination was performed due to the absence of blood flow in both jugular veins observed during ultrasound, and confirmed a thrombus in the CrVC (30). In previous studies, findings such as a focal filling defect in the dorsal aspect of the CrVC or a mixed texture of gas and soft tissue instead of contrast media in the CrVC were identified on CTA (11,30). In this case, a filling defect of approximately 3 cm in length within the CrVC was identified on CTA, and it occupied most of the lumen. Therefore, a portion of the contrast medium was trapped cranially.

In a previous study, the poor prognosis for dogs exhibiting clinical signs associated with CrVC thrombosis was reported (31). Out of 17 dogs with clinical signs associated with CrVC thrombosis, 15 either died or were euthanized within 20 days (31). One of the two surviving dogs was treated with heparin, an anticoagulant agent, while the other received streptokinase, a thrombolytic agent (31). In veterinary medicine, there are few reports on the use of thrombolytic agents. The use of rt-PA for treating CrVC thrombosis is particularly limited in dogs. In recent studies, there have been reported cases in which the application of rt-PA in patients with CrVC thrombosis resulted in a reduction in the size of the thrombus, leading to clinical improvement and survival, followed by discharge. Bleeding is a common complication in patients after the administration of rt-PA (5,11,12). However, in some cases, despite the application of rt-PA, the condition worsened, resulting in death or euthanasia (27,30). In this case, rt-PA was administered 1 month after the diagnosis of CrVC thrombosis, and no bleeding was observed. A follow-up CT scan 3 months later confirmed a reduction in the size of the thrombus. Although a very small amount of chylous effusion persisted due to residual thrombus, the patient survived for 18 months after the diagnosis of CrVC thrombosis, but eventually died due to a trauma unrelated to thrombosis. The guidelines for administering rt-PA in veterinary medicine have not been established yet; however, according to one study, the administration of rt-PA in dogs has been demonstrated to be safe and effective (12).

Although CrVC thrombosis is an uncommon presentation in veterinary patients, CrVC thrombosis should be considered as a differential diagnosis of chylothorax. CT lymphangiography can help confirm or rule out chylothorax caused by the rupture or congenital anomaly of the thoracic duct. CTA can complement the limitations of CrVC thrombosis diagnosis through ultrasound examinations and assess potential sites, beyond the CrVC, of thrombus formation.

This study was supported by a grant from the Cooperative Research Program for Agriculture Science and Technology Development (Project No. RS-2023-00231792), RDA, Republic of Korea.

The authors have no conflicting interests.