Magnetic resonance imaging and pathological findings of aortoiliac thromboembolism with a thoracic aortic mural thrombus in a Japanese Black calf

Abstract

A 24-day-old female Japanese Black calf presented a sudden paraplegia after a history of watery diarrhea. Antemortem magnetic resonance imaging confirmed the suspicion of thrombotic component in the abdominal aorta, without any spinal cord abnormality at the lumbar region. On necropsy, a massive thrombus occupied the lumen from the distal abdominal aorta to the bifurcation of the external iliac arteries. In the thoracic aorta, another mural thrombus developed from the caudal side of the incompletely closed ductus arteriosus orifice, with aortic wall erosion. Both thrombi were mainly composed of platelets. Any microbes were undetected during organ and thrombus incubations. A saddle embolism in the abdominal aorta occurred by an abacterial white thrombus suspiciously originated from the thoracic aortic mural thrombus.

Several previous veterinary studies have described involvements of arterial thrombosis in various species [1, 3,4,5, 7, 8, 10, 12, 16,17,18,19,20,21,22]. Arterial thrombosis in the aorta can occur embolism found at typically the aortoiliac bifurcation with downstream ischemia and hindlimb paralysis [1, 3, 4, 12, 16,17,18, 22]. Various pathogenic factors of arterial thrombosis included hypertrophic cardiomyopathy in cats [10] and protein-losing nephropathy, neoplasia, exogenous corticosteroids, or infection in dogs [21]. Arterial thrombosis may be caused in horses by excessive exercise especially in thoroughbred [18] or inflammation affecting the mesenteric artery in cases of strongylosis [17]. In cattle, arterial thrombosis has been sporadically documented in varying ages, breeds, and countries; it can occur secondary to severe diarrhea [4, 7, 16, 19] and/or systemic bacterial infection [1, 5, 7, 8, 12, 20]. In veterinary practice, ultrasonography and contrast-enhanced computed tomography (CT) were diagnostic imaging modalities contributing to the antemortem detection of aortoiliac thromboembolism in the previous bovine cases [4, 12]. Magnetic resonance imaging (MRI) also helps to detect thrombus lesions as reported in human medicine [2] and canine patient with aortoiliac thromboembolism [3, 22]. However, MRI has not been used to diagnose this disease in cattle. Aortic mural thrombosis, another type of arterial disease from aortoiliac thromboembolism, is a firmly rare disease in veterinary medicine characterized as thrombus formation attachment to the aortic wall; it typically occurred due to aneurysms and arteriosclerosis [14, 15] or, although rare, in normal or minimally atherosclerotic aorta secondary to conditions resulting in hypercoagulable states including endothelial (or intimal) injury, congenital coagulation disorder, and malignant tumors as reported human patients [9, 23]. This report aimed to show the diagnostic efficacy of MRI to detect aortoiliac thromboembolism and to assess the disease pathogenesis, which was accompanied with thoracic aortic mural thrombosis, based on macroscopic and pathological findings.

A 24-day-old female Japanese Black calf with a history of watery diarrhea from a few days after birth, followed by collapse in recumbency with severe dehydration and hypothermia at 10 days of age. After intravenous infusion and oral rehydration treatments, the calf recovered enough to stand up once but suddenly showed astasia at 14 days of age. On admission, the calf weighed 48 kg and presented with paraplegia in its hindlimbs, despite having the ability to move its forelimbs, resulting in wriggling and positioning of sternal recumbency (Fig. 1). The muscular volume was atrophic in bilateral hindlimbs, in which thigh region thinning was more severe. Furthermore, alopecia and decubitus ulcer lesions were macroscopically detected in the whole skins of both hindlimbs, especially in the hocks, buttocks, and caudal side of upper thighs. Palpation of both hindlimbs demonstrated pulseless femoral arteries. The distal skin and claw areas were also cold on palpation. Vigor, appetite, and other vital signs were normal. Neurological examination identified a lower motor neuron sign such as complete losses of patella tendon and flexion reflexes. Deep pain sensation was also not observed in both hindlimbs. Blood examination revealed leukocytosis and thrombocytosis; blood biochemistry showed increased serum liver, muscle and bone enzymes (Table 1). For diagnostic imaging examinations using a 16-section multidetector CT scanner (ECLOS; Hitachi Co., Ltd., Tokyo, Japan) and a low-field MRI machine (AIRIS Vento 0.3 T; Hitachi Medical Corp., Tokyo, Japan), the calf was premedicated with intravenous xylazine (0.2 mg/kg; Celactal, Bayer Yakuhin, Ltd., Osaka, Japan) and intubated and then anesthetized with isoflurane inhalation (2–3%; ds isoflurane, DS Pharma Animal Health, Osaka, Japan). On CTs obtained when scanning the head, chest, and abdomen, no skeletal abnormalities were detected in the entire vertebrae. MRI revealed no spinal cord abnormalities within the thoracolumbar vertebrae and the cauda equina. In sagittal MRI sections, a massive thrombotic component was observed within the abdominal aortic lumen at the lower lumbar vertebrae (L5–L6) level with moderate to high signal on T1-weighted images and low signal on T2-weighted images (Fig. 2). Compared with the posterior vena cava in transverse MRIs, the signal intensity of the thrombus is slightly higher on T2-weighted images; however, it is depicted with an even higher signal on T1-weighted images. The clinical signs and diagnostic results indicated an aortoiliac thromboembolism in this case. The calf was considered to have poor prognosis because of the MRI findings of the extended length of thrombotic component within the abdominal aorta and severe, progressive weakness in both hindlimbs. Thus, euthanasia was selected from the viewpoint of animal welfare and humane care and was performed by exsanguination under the unconscious state obtained by intravenous administration of pentobarbital (50 mg/kg, Somnopentyl, Kyoritsu Seiyaku Corp., Tokyo, Japan), according to the euthanasia guidelines [13].

Fig. 1.

Posture of a 24-day-old calf presenting with paraplegia. Alopecia and decubitus ulcer lesions are evident in the atrophic pelvic and thigh regions in the hindlimbs.

Table 1. Hematological and serum biochemical values in a 24-day-old calf with paraplegia.

Variables	Results	Reference range [1, 25]
Erythrocytes (×104/µL)	794	510–760
Hemoglobin (g/dL)	10.4	8.5–12.2
Hematocrit (%)	30.5	22–33
White blood cells (×103/µL)	22.7	4.9–12
Platelet count (×104/µL)	164	19.3–63.7
Blood urea nitrogen (mg/dL)	11.6	10–25
Creatinine (mg/dL)	0.3	0.90 ± 0.19
Alkaline phosphatase (IU/L)	487	27–107
Aspartate transaminase (IU/L)	263	37.4 ± 4.3
γ-glutamyltranspeptidase (IU/L)	77	481.1 ± 318.3
Creatine kinase (IU/L)	661	4.8–12.1

Fig. 2.

Sagittal and transverse T1-weighted (A and B, respectively) and T2-weighted planes (C and D, respectively) of caudal abdominal region. Sagittal T1- and T2-weighted planes indicate the partial length of the abdominal aorta’s lumen including a massive thrombotic component (arrowheads) represented by moderate to high and low signal intensities, respectively. On these planes, the spinal cord (SC) structure appears to be normal. On transverse T1-weighted and T2-weighted planes at the sixth lumbar vertebra (L6), the thrombotic component (arrowheads) appears as higher and slightly higher signal intensities compared with those of the posterior vena cava (VC), respectively. Bars=2 cm.

During necropsy, the muscular structures were diffusely dried and faded in the whole areas of both hindlimbs. Arterioles were necrotic in the cut surface of hindlimbs. A gray-brown, elastic thrombus measuring 4 cm in length occupied the vascular lumen of the distal abdominal aorta up to its bifurcation into the external iliac arteries (Fig. 3A). In the thoracic aorta, a white to pink, firm, irregularly surfaced mural thrombus was developed from the caudal side of the ductus arteriosus bifurcation, covering the ductus arteriosus orifice (Fig. 3B). The ductus arteriosus lumen was occupied by a bloody to white, jelly-like substance. The ruminoreticulum mucosa was poorly developed, and its content was scanty. The jejunal mucosa was thinned. The right and left anterior lobes of the lungs were locally hepatized. No significant findings were detected in other organs. Fresh samples originated from the heart, liver, lungs, kidneys, spleen, and thrombus were incubated both aerobically and anaerobically on the sheep blood agar medium (37°C, 3 days). These samples were also incubated on the potato dextrose agar medium with chloramphenicol (room temperature, 7 days) for fungal isolation. However, any microbes were not detected in these incubations.

Fig. 3.

Macroscopic views of the aortic thrombi detected in a 24-day-old calf. (A) Dorsal incision of the distal abdominal aorta (AAo). A massive thrombus (arrowheads) adheres to the aortic lumen. (B) In the thoracic aorta (TAo), a mural thrombus protrudes from the caudal side of the orifice of ductus arteriosus (arrow). AA, aortic arch.

Necropsy samples were fixed in 15% phosphate-buffered neutral formalin, embedded in paraffin, and sectioned at 3 μm. Sections were treated with hematoxylin and eosin (HE). Selected sections were additionally stained with Azan, Victoria blue-HE, phosphotungstic acid hematoxylin (PTAH), periodic acid-Schiff (PAS), Grocott’s methenamine silver, Ziehl–Neelsen, and Gram staining. Microscopically, the thrombus that embolized the abdominal aorta consisted of rich eosinophilic granules (Fig. 4) and few meshed fibrin, erythrocytes, leukocytes, and cellular debris, with partial organization and recanalization of the small blood vessels; the predominant granules were considered as platelets because of negative PTAH and PAS staining and colorless to pale bluish purple by Azan staining. The abdominal aorta lacked any marked erosion, necrosis, or other pathological changes of the aortic wall. The iliac arteries distal to the embolism site contained erythrocyte- and fibrin-rich thrombus, associated with diffuse arterial wall necrosis. The thoracic aortic wall thrombus was attached to the aortic intima with erosion and partial degeneration of the elastic fibers of the media (Fig. 5). The mural thrombus was mainly composed of platelets (white thrombus), consistent with the thrombus embolized the abdominal aorta. The ductus arteriosus lumen measured approximately 2 mm on the tissue section. The red thrombus inside the ductus arteriosus had a layered structure consisting of areas rich and poor in erythrocytes (lines of Zahn). In the skeletal muscles of the hindlimbs, the muscle fibers were diffusely swollen, lacked striations, and revealed marked eosinophilia (hyaline degeneration) with dystrophic calcification, hyaline thrombus formation in small blood vessels, and inflammatory cell infiltration, mainly neutrophils, into the muscle interstitium. The skeletal muscle degeneration was accompanied by multifocal fibrinopurulent dermatitis. In the jejunum, catarrhal enteritis with shortened villi was detected. Multiple central chromatolysis was observed in large motor neurons at the bilateral ventral horns in the lumbar spinal cord. No other abnormalities were found in major organs including the heart.

Fig. 4.

Histopathological findings of a thrombus within the lumen of the caudal abdominal aorta. (A) The thrombus structures are present fully within its lumen with partial organization (arrowheads) and recanalization of the small blood vessels (arrow). Hematoxylin and eosin (HE) stain. (B) During the bifurcation of the abdominal aorta and bilateral external iliac arteries (EIA), the thrombus (T) structures are mounted toward the EIA’s lumens. HE stain. (C, D) In a magnified view of the square box in image A, the thrombus reveals eosinophilic granules (platelets) predominantly, with few meshed fibrins, erythrocytes, leukocytes, and cellular debris. HE (C) and Phosphotungstic acid hematoxylin (D) stains. Bars=20 μm.

Fig. 5.

Histopathological findings of a thoracic aortic mural thrombus. (A) The thrombus that cut longitudinally through the center of the ductus arteriosus lumen (asterisk) was attached to the aortic wall. Hematoxylin and eosin (HE) stain. (B) In a magnified view of the square box in image A, the thrombus (T) is directly attached to the aortic media (M) where the intima (I) disappears (arrowheads). Victoria blue-HE stains.

The present case revealed thromboembolism at the distal abdominal aorta and external iliac arteries, resulting in hindlimb ischemia and necrosis. Regarding the diagnostic imaging techniques contributing to the antemortem disease diagnosis, ultrasonography [4] and contrast-enhanced CT [12] were effectively used in previous reports. The CT scan with intravenous injection of a contrast material can be considered a strong indicative of empty blood flow with thrombus lesion, but no contrast enhancement is evident within the vessel’s lumens located caudally or distally than the lesion [12]. Although it would be useful for detecting the pathological changes in large vessel shapes and sizes, plain CT could not provide effective significant evidence of the intra-arterial thrombus in the present case. Conversely, based on the present results, MRI effectively demonstrated a successful aortoiliac thromboembolism without using a contrast agent. To our best knowledge, this is the first bovine case report demonstrating the antemortem use of MRI facilitating to identification of aortoiliac thromboembolism. Furthermore, MRI can simultaneously evaluate the spinal cord and the abdominal aorta on the same images of the caudal abdominal region, as identifying no abnormal spinal change and the intra-arterial lesion in the present case. This finding indicates the diagnostic efficacy of MRI to differentiate this disease from other spinal and vertebral diseases, when used alone. The high signal intensity of the aortic thrombus on T1-weighted images correspond with its time course reported on a previous study in swine model; on both T1- and T2-weigted images, the signal intensity was high 1 week after affection and then gradually decreased until turned very low within the next several weeks [6]. Thrombus-age-related MRI changes using the canine model were characterized by the shortened T1 and T2 relaxation times dependent on increase in thrombus ages, causing the high and low signal intensities in T1- and T2-weighted images, respectively [22]. In clinical use of MRI for the previous canine case, aortic thromboembolism was represented by presence of the heterogenous thrombus contents within the aorta lumen [3]. Despite the unclear reasons behind the thrombus showing low signal intensity on T2-weighted images in the present results, it might have been influenced the differences of imaging area, pathogenesis, thrombus age, and composition such as the degree of organization and erythrocyte inclusion.

The etiological factors contributing to thrombus formation included vascular wall lesions, abnormal blood flow, and hypercoagulable state such as systemic bacterial infections [16]. A previous bovine report of this disease has confirmed that the thrombus was primarily composed of fibrin and was attributed to sepsis by Pseudomonas aeruginosa and Alcaligenes spp. infections [1]. Another report describing a 6-month-old Charolais calf with the iliac artery embolism also confirmed a red thrombus, mainly comprising fibrin, originated from the left atrioventricular valve due to endocarditis associated with systemic Staphylococcus aureus infection [20]. Similarly, other studies also have reported accompanying infections [5, 7, 8, 12]. In contrast, the embolic thrombus in the present case was a white thrombus (platelet thrombus) without evidence of bacterial infection. The discrepancy in thrombus properties between the current and previous reports may reflect differences in the thrombus formation mechanisms, which is unusual in that this case was not caused by infection. Severe diarrhea was observed before the onset of aortoiliac thromboembolism, consistent with previous reports of thrombosis in calves [4, 7, 16, 19]. Dehydration and endotoxemia associated with diarrhea might contribute to hypercoagulability. Unfortunately, we did not conducted pathogen testing to diagnose the cause of diarrhea because it had been cured for more than a week at the time of our examination; thus, we could not identify the source of the preceding diarrhea in the present case. Thrombocytosis is known to occur as a reactive change in several conditions such as infections, surgery, neoplasia and/or inflammation and does not result in increased risk of thrombosis [11]; therefore, we considered that this finding observed in the present case is a nonspecific, secondary finding associated with hindlimb necrosis and is considered not to be the pre-existing cause of thrombus formation. Other potential causes of a hypercoagulable state, such as congenital coagulation disorders, could not be evaluated in the present case.

We speculated that the patent ductus arteriosus is one of the causes of the mural thrombus because it can generate turbulent flow along the bifurcation between ductus and the descending aorta, increasing shear stress and subsequent intimal damage. This intimal injury can trigger the release of tissue factors from the vessel wall and various platelet-activating factors, resulting in localized platelet aggregation and coagulation system activation as the cause for white thrombus formation [24]. This theory is supported by the present observation of a mural thrombus formation on the caudal side of the orifice of ductus arteriosus, along with aortic wall erosion. Furthermore, in the present case, pathological changes of the aortic wall were rarely observed in the abdominal aorta in contrast to the thoracic aorta, even though both thrombi were histopathologicallly identical. In human medicine, there are known cases where wall-attached or free thrombi in the thoracic aorta have caused peripheral embolism in the mesenteric arteries or lower limb arteries [15]. From the above, it is speculated that the embolic thrombus in the aortoiliac artery may have originate from a mural thrombus formed initially in the thoracic aorta, which then detached and reached the downstream aorta. The limitation of this report is the lack of information regarding therapeutic interventions for thrombosis in the calves, which is expected to be a significant challenge in future similar cases.

In conclusion, this case was diagnosed with a saddle embolism at the aortoiliac bifurcation by an abacterial white thrombus in the calf, which might originate from a mural thrombus in the thoracic aorta. This case highlights efficacies of antemortem MRI on diagnosis of aortic thromboembolism and importance to considering thromboembolism in the differential diagnosis for hindlimb paralysis, even without signs of systemic infection in calves.

CONFLICT OF INTEREST

There were no declarations or conflict of interest.