Feline Cerebrovascular Disease: Clinical and Histopathologic Findings in 16 Cats

Introduction

Cerebrovascular disease (CVD) is currently topical in the veterinary literature and much information has been recorded for the dog.^1–4 Cerebellar infarcts have been described in two cats; however, data on feline vascular encephalopathies are limited.⁵ Vascularization in the feline brain is different than in dogs rendering differences in lesion distribution feasible. The cerebral arterial circle of Willis on the ventral surface of the brain collects blood from contributing arteries. The function of the arterial circle is to maintain a constant blood pressure in the end arteries.⁶ In the cat, the arterial circle is supplied by anastomoses with the maxillary and pharyngeal arteries, which arise from the external carotid artery. The proximal portion of the internal carotid artery, which is the main supplying vessel in the dog, becomes obliterated in cats soon after birth. The maxillary artery supplies the arterial circle by an anastomosing ramus, which includes a rete mirabile (i.e., a network of fine arterioles). The arterial circle in the cat is not a closed ring because of the lack of a rostral communicating artery that is normally present in other species. Also, the direction of flow in the basilar artery is away from the arterial circle so that maxillary blood from the external carotid arteries is distributed to the entire brain, except for the caudal portion of the brainstem, which is supplied by the vertebral arteries (Figure 1).^7–11

View Figure

• Download as Powerpoint
• Download Figure

CVD refers to any abnormality in the brain resulting from pathology of the supplying blood vessels. Insufficient blood supply leads to ischemia or infarction in which normal cellular function cannot be maintained.⁶ Severe ischemia results in an area of necrosis, termed malacia, ischemic necrosis, or infarction.⁶ Ischemia results from arterial or venous obstruction caused by emboli or thrombi whereas rupture of a vessel leads to hemorrhage.^1,2,6,12 A cerebrovascular accident may be either ischemic (pale or bland), in which the cellular reactions to ischemic necrosis predominate, or hemorrhagic (red) infarcts, wherein the lesions are associated with hemorrhagic phenomena. An important pathogenetic aspect of hemorrhagic infarction is the fact that tissue reperfusion or oxidative injury can be sequels of mobilization of the thrombus. Bleeding may occur in a capillary bed damaged by the ischemic insult.¹³

Intraparenchymal hemorrhage in humans occurs in a variety of circumstances, which are in decreasing frequency as follows: hypertension, cerebral amyloid angiopathy, anticoagulant administration, primary or secondary brain neoplasms, arteriovenous malformations, aneurysms, or drug abuse.¹³ Neuronal damage occurs in intracranial hemorrhage due to compression and distortion of tissue.⁶ Cerebrovascular accidents, or strokes, are the most common clinical presentation of CVD.⁶

In the current study, a review of the literature on cerebrovascular disease in cats was carried out. A variety of causes of ischemic CVD in cats was identified including cuterebra parasitic migration, migrating heartworms, anesthesia-related ischemia, intracranial telangiectasia, granulomatous meningoencephalitis, and experimental- and food-related thiamine deficiency.^14–23 In a proportion of cases the cause of the cerebrovascular event remained undetermined.¹⁵ Intraparenchymal hemorrhage in cats has been associated with primary or secondary hypertension, cerebral amyloid angiopathy, and intracranial neoplasia.^24–27 To the authors’ knowledge, no retrospective study describing clinical and histologic findings in a larger population of cats with CVD has been published in the veterinary literature. The purposes of this study were to describe clinical and histopathologic findings in cats with CVD and to identify possible associations with concurrent disorders in a nonrandomly selected population of cats.

Materials and Methods

The medical records of cats with a histopathologic diagnosis of CVD of nontraumatic origin recorded at the Department of Pathology at the University of Zurich between 1998 and 2004 were reviewed. Originally, 22 cats with complete medical records were identified. Of these, five were excluded due to the presence of meningioma and the likelihood of the tumors contributing to the clinical signs. In addition, the only cat affected with an intraventricular hemorrhage was removed from the study. In this cat, a ventricular hemorrhage was followed by an infarct due to an increase in intracranial pressure. Results of full postmortem examinations were available in 15 of the 16 cats. In one cat (case 8), an ischemic infarction was diagnosed by computed tomography (CT)-guided biopsy. This cat was still alive at the time of submission of this paper.

Diagnostic tests included a hematologic and biochemical profile in all cats. Thoracic and abdominal radiographs were obtained in five cases (cases 1, 2, 4, 5, 12). Specific biochemical tests for evaluation of thiamine status or coagulopathies, functional liver tests, and measurement of blood pressure were not available in this retrospective study. Neurodiagnostic procedures included cerebrospinal fluid (CSF) analysis in two cats (cases 3, 8) and CT of the brain in two cats (cases 8, 12).

Results

A total of 16 cases met the inclusion criteria. Two purebred cats were included (Karthäuser, Persian), 10 were males (six neutered) and six were females (four neutered). Based on the results of histopathology, the cats were categorized as having suffered from either an infarction (n=12) or a hemorrhage (n=4). Ischemic infarction was found in seven cats (five male and two female) and red infarcts in five cats (two male and three female). Intracranial hemorrhage was shown on postmortem in four cats (three males and one female).

The median age of all 16 cats was 8.5 yr (range, 2–21 yr). Cats with either a red or pale infarction were of a median age of 8 yr (range, 2–16 yr), whereas four cats with an intracranial hemorrhage had a median age of 9.5 years (range, 3–17 yr).

Ten of the 16 cats showed a combination of neurologic and extraneurologic signs (urinary incontinence included). In six of the 16 cats only neurologic signs were found. The results of the neurologic examinations are shown in Tables 1A and B. The described clinical signs might have been the result of disorders of noncerebrovascular origin (Tables 2A and B).

Table 1A Historical Complaints and Clinical Findings for 12 Cats with Ischemic Cerebrovascular Disease

View Fullscreen

Table 1B Historical Complaints and Clinical Findings for Four Cats with Intracranial Hemorrhage

View Fullscreen

Table 2A Findings in 12 Cats with Infarction

*

Case 8 had a brain biopsy with full recovery. All other cases, 1–7 and 9–12, had postmortem examinations.

G, grossly visible; M, microscopic

View Fullscreen

Table 2B Morphologic Findings in Four Cats with Intracranial Hemorrhage and One Cat with Intraventricular Hemorrhage

FIP, feline infectious peritonitis

View Fullscreen

In 11 of the 12 cats with an infarction, an acute onset of clinical signs was seen. One cat (case 3) with meningoencephalitis and gastrointestinal lymphoma showed clinical signs for 1 mo prior to presentation. All of the other cats were presented to the Division of Small Animal Surgery of the University of Zurich, Switzerland, within 10 days of first developing clinical signs. The clinical signs were sudden in onset in cats with intracranial hemorrhage.

Neurologic signs showed a correlation with the localization of the cerebrovascular lesions in all cats. Underlying conditions found on postmortem examination (Tables 2A and B) could have contributed to any of the clinical signs. In the five cats with an infarction of the forebrain, the neurologic signs included central blindness, seizures, altered state of consciousness, proprioceptive deficits, and circling. In the five cats with an infarction in the brainstem, the neurologic signs included an altered state of consciousness, absent menace responses, head tilt, ataxia, tetraparesis, and anisocoria. Case 6 had an ischemic infarction in the basal nuclei and demonstrated an absent menace response, paresis of the left hind limb, and loss of pupillary light reflexes. In case 10, a cat with an ischemic infarction of the brainstem and the thalamus, ataxia was reported. Clinical data are summarized in Tables 1A and B.

Case 15, a cat with brainstem hemorrhage, died suddenly. Two cats with intracranial hemorrhage did not recover from status epilepticus (Table 1B), and case 13 showed tetraparesis and extraneurologic signs over 10 days (Tables 1B and 2B).

Asymmetric or lateralized clinical signs were noticed in four of the 12 cats with an infarction. These signs included hemiparesis, anisocoria, unilateral menace response deficit, unilateral paresis, and head tilt (cases 3, 6, 9, 11).

The extraneurologic clinical signs are outlined in Tables 2A and B. In the 12 cats with an infarction, inappetence (n=3), lethargy (n=3), dyspnea (n=2), fever (n=2), cardiorespiratory arrest during anesthesia (n=1), weight loss (n=1), inability to jump (n=1), diarrhea (n=1), vomitus (n=1), absent peripheral pulses (n=1), incontinence (n=1), and vocalization (n=1) were described. Two of the five cats with a hemorrhagic infarction were icteric. Thoracic radiographs (performed in cases 1, 2, 5, 12) confirmed thoracic effusion in two cats with pyothorax, lung edema in a cat with cardiac failure, and a diffuse infiltrative pulmonary pattern in one cat with severe lipid pneumonia.

The abnormal laboratory findings corresponded to the underlying metabolic disease (e.g., icterus with liver disease). The T₄ level was high or the thyroid stimulating hormone tests were positive in three cats (cases 1, 4, 13). In two cats, CSF was obtained by cisternal puncture. Mononuclear pleocytosis was found in case 3 and a mild meningoencephalitis was confirmed postmortem. A mixed cellular inflammation with albuminocytological dissociation was seen in case 8. This was compatible with tissue destruction through infarct. The CT-guided biopsy sample obtained in this case showed no evidence of inflammation.

CT was used to localize a lesion in the cerebral cortex of case 8. To determine the identity of the lesion, surgical biopsy by a minimal invasive approach was carried out and an ischemic infarction diagnosed. The cat made a full recovery. In case 12, the lesion was not visible on CT. This cat's hemorrhagic infarction was localized in the brainstem.

A full postmortem examination was performed on 15 cats (Tables 2A, B, and Table 3). Gross lesions were visible in four of the 15 cats (Figure 2). In two of the cats with ischemic CVD, abnormal tissue was grossly visible and in one cat with red infarcts petechial bleeding was seen. Gross lesions were obvious in one cat with intraparenchymal hemorrhage and changes found in the histopathology are shown in Figures 3 and 4.

Table 3 Classification of Stroke

View Fullscreen

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

Infarction and malacia were seen in the cerebral cortex of three of the 12 cats with an infarction, in the brainstem of five of the 12 cats with an infarction, in both the cerebral cortex and brainstem in three of the 12 cats with an infarction, and in the basal nuclei of one cat with an infarction. All the red infarcts were localized to the brainstem, whereas pale infarcts and malacia were seen in the forebrain, brainstem, hippocampus, and basal nuclei. Cases were graded as either large vessel disease (i.e., territorial infarcts) or small vessel disease (i.e., lacunar infarcts). Of the 12 cats with an infarction, territorial infarcts were seen in three cats, lacunar infarcts in seven cats, and global ischemia in two cats (Table 3). Intraparenchymal hemorrhage was multifocal in two cats and focal in two cats (Table 2B).

Extraneurologic clinical signs of CVD were correlated with the following potentially predisposing conditions, which were confirmed by pathologic examination in 15 out of 16 cats: hyperthyroidism, heart disease, nephropathies, and hepatopathies. Meningitis was a feature in five of these 16 cats. The inflammation was mononuclear in case 6 and mixed cellular in cases 3 and 8. In cases 13 and 14, the inflammation was seen to be attributed to feline coronavirus on postmortem.

Concurrent diseases were noted on postmortem examination in 14 of 16 cats (Tables 2A and B). All cats with red infarcts (n=5) showed severe liver pathology and two of these cats were icteric. Overall, seven of 16 cats had liver disease on postmortem examination. Chronic interstitial nephritis was a feature in three of the 16 cats. Three cats had pathologic thyroid changes. Thyroid adenoma was found in two of these cats and thyroiditis in one cat. All three cats had confirmed positive results for thyroid disease. Two cats were originally treated for pyothorax and pleuritis and mediastinitis were identified on the postmortem examinations. One of these cats suffered cardiorespiratory arrest during thoracocentesis and the other showed signs of dyspnea and hypoxia prior to death. Ischemic polioencephalomalacia was diagnosed in both of these cats postmortem. One cat died of cardiac failure (case 5). One cat with leukoencephalomalacia was found to have gastrointestinal lymphoma and mild meningoencephalitis (case 3).

The findings in cats with intracranial hemorrhage are summarized in Table 2B. Meningitis was an intracranial finding in two cats and changes in one of these two cats were associated with feline coronavirus infection. Meningitis was a likely cause of the hemorrhage in both cases. Case 15 suffocated during recovery from anesthesia for ocular surgery. No concurrent disease was determined in case 16.

In the absence of concurrent abnormalities, cerebrovascular lesions were cryptogenic in two of 16 cats. This includes the one cat that recovered fully after CT-guided biopsy.

Discussion

Information about cerebrovascular disease in cats is limited. The medical histories of 16 cats with cerebrovascular disease that were extracted from a computerized pathology database were analyzed. One cat in which a CT-guided biopsy revealed an ischemic infarct was included in the study. The infarcts were found to be hemorrhagic in five cats and ischemic in seven cases. The other four cats suffered intraparenchymal hemorrhage.

Publications describing the unique intracranial blood supply of the cat's brain are sparse. In the cat, the blood supply of the brain is from one main source: the external carotid artery. Only the caudal brainstem is supplied by the vertebral arteries, and cats have a rete mirabile, a structure that is not present in dogs.^9,10

Cerebellar infarcts in Cavalier King Charles spaniels have been reported with increasing frequency, and recently, two cases of cats with cerebellar infarcts have also been reported.^2,5 In the current study, a cerebellar infarct was not identified in any cat. Instead, the brainstem was affected in seven cats. An ischemic event during anesthesia can cause neuronal cell loss; however, cells in the cerebellum are considered more resistant to ischemic insults.²⁰

Several diseases or factors have been associated with cerebrovascular disease in cats: cuterebra parasitic migration, migrating heart worms, anesthesia-related ischemia, intracranial telangiectasia, granulomatous meningoencephalitis, and thiamine deficiency.^14–23 Intracranial hemorrhage has been associated with primary and secondary hypertension, cerebral amyloid angiopathy, and intracranial neoplasia.^24–27

Stroke is the sudden and abrupt onset of focal neurologic deficits from an intracranial vascular event.⁶ Only one cat with underlying meningitis showed development of clinical signs over 1 mo, indicating that signs of cerebrovascular disease can be obscured by concurrent disease.²⁸

Neurologic signs, seen in six of the 16 cats included in this study, were consistent with the localization of the cerebrovascular accident (Tables 1A and B) as reported elsewhere.^5,14,15 Ischemic lesions were most commonly associated with seizures as the major neurologic sign. Case 7, with a focal cortical ischemic lesion, presented with refractory seizures as the only clinical sign. In the literature, only a few reports mention seizures as a clinical sign in cats with CVD.^23,24 Central vestibular signs were associated with brainstem involvement. Asymmetric neurologic signs were noted in four of 12 cats with an infarction.

Extraneurologic clinical signs were seen in 10 of the 16 cats. They could be nonspecific or due to underlying disease as shown at postmortem (Tables 2A and B). Inappetence was seen in three of the 12 cats with infarction and icterus in two of the five cats with hemorrhagic infarcts. Cats with cerebrovascular disease may be presented with neurologic signs or extraneurologic signs only.

Hypercoaguable states were proposed by several authors as a predisposing factor for CVD in humans and dogs.^1,2,23 Tests for coagulation abnormalities were not carried out in the cats featured in this study and so coagulopathies could theoretically have been missed. In the literature, the clinical signs of feline coagulopathies include hematuria, cutaneous hematomas, gingival bleeding, epistaxis, or prolonged bleeding times.^29–32 None of these signs were recorded for any of the cats in the current study; however, liver disease (e.g., hepatic lipidosis) was the most commonly cited cause of coagulation disorders.^30,33 Indeed, seven cats in the current study were diagnosed as having hepatic lipidosis, hepatitis, or liver telangiectasia. Thus, a possible predisposition for CVD cannot be ruled out, particularly in cats with hemorrhagic infarction (Tables 2A and B). In fact, kidney disease, feline infectious peritonitis, lipidosis, lymphoma, and hypertrophic cardiomyopathy are all very common findings. Hyperthyroidism is common, but lipid pneumonia and meningoencephalitis are rare findings at general necropsies.^28,34

In three cats in this study (cases 10, 11, 12), histopathologic findings of the brain resembled those seen in thiamine deficiency. They consisted of hemorrhages in the caudal brainstem and pontine nuclei affecting spinoreticular tracts in a symmetric fashion.²¹ Thiamine deficiency is a primary metabolic disease, but it can result in vascular lesions induced by cerebral hypoperfusion of vulnerable structures, regionally selective neurodegeneration, and generalized impairment of oxidative metabolism.^35–37 The histologic abnormalities found in these cases do not prove the presence of thiamine deficiency because the end pathologic process of hypoxia is similar. Hypoxia was a feature in four of the 16 cats in the current study. Antemortem measurements of the absolute concentration of thiamine in whole blood or determination of erythrocyte transketolase activity would have been necessary to give evidence that these cats were really deficient in thiamine.

The number of cats originally retrieved with a diagnosis of CVD of nontraumatic origin represents approximately 3% of brain necropsies conducted in cats at the University of Zurich during the 6 yr period of this study; however, the true incidence of CVD in cats is suspected to be higher as this figure does not include cases that were not euthanized. Therefore, no conclusions may be made about the real incidence of CVD in cats. The stroke prevalence in humans is approximately 5% in men and 2.5% in women, with the highest risk between the ages of 65 and 74.^38,39 The current study identified a similar trend as older cats were more often affected because the median age was 8 yr, but the age range was wide (2–17 yr).

In humans, higher age prevalence is explained by the incidence of chronic, progressive atherosclerosis, a disorder that is very rare in cats and was not seen in any cats included in this study.⁴⁰ The higher age prevalence of CVD in cats could be related to the occurrence of various geriatric diseases, but it was difficult to prove their association with CVD. Renal disease and hyperthyroidism are common in older cats and were each diagnosed in 3 of the 16 cats. Therefore, CVD as a feature of renal- or hyperthyroid-induced hypertensive encephalopathy would represent a possible link between these disorders.^24,41 Blood pressure measurements were not available in the cats reviewed in this study, and lesions characteristic of vascular disease were not present in the histopathologic specimens in any cat. The pathologic changes in the kidneys were chronic and of slight to moderate severity. In two cats, these were severe, chronic interstitial and fibrosing (case 4, 16), but no changes in the vessel walls or intima were seen histologically in the kidneys and the brain (i.e., findings that are diagnostic hallmarks of hypertensive encephalopathy).⁴¹

Hemorrhage can be due to damage to the vascular endothelium or vessel rupture, especially when perfusion is partially restored (Figure 5).⁴² Endothelial damage has been associated with vasculitis, amyloid angiopathy (amyloid deposits within vessel walls), or hypertension (arteriosclerotic microvascular changes).¹³ In cases 13 and 14, vascular and inflammatory changes were found, which were attributed to feline infectious peritonitis. In this regard, the value of CSF examination is highlighted. Although it is not specific, CSF analysis is a sensitive test that may provide valuable additional information (e.g., content of specific cell types) about associated diseases, such as inflammation or neoplasia, thus being helpful for further diagnostic or therapeutic decisions.²⁸

View Figure

• Download as Powerpoint
• Download Figure

CT was able to locate the site of the CVD in case 8, but not in case 12, which suffered hemorrhagic infarction of the brainstem. There is difficulty evaluating neural structures within the caudal fossa using CT due to beam-hardening artifact.^3,43 This may have been the reason why CVD could not be detected in case 12 by CT imaging. CT and magnetic resonance imaging (MRI) can be useful in obtaining minimally invasive biopsies. In general, MRI is more sensitive for the detection of vascular encephalopathies. Recent advances in MRI technology mean that even in the presence of hemorrhage in early stroke, CT offers no advantage.² One argument that may justify the use of CT in patients with suspected CVD of nontraumatic origin is that CT may exclude possible diagnoses and therefore decrease the anesthetic risk for the veterinary patient due to the shorter scanning time, although this will depend on available equipment.

Conclusion

The combination of individual neurologic and extraneurologic clinical signs is influenced by the type of CVD, the localization of the intracranial lesions, and any underlying pathology. Hemorrhagic infarcts should be suspected in cats with liver disease and central vestibular signs. The coexistence of hyperthyroidism, coagulation disorders, or thiamine deficiency could not be excluded in the cats reviewed in this study. Future studies using measurements of blood pressure, liver profiles, coagulation profiles, and thiamine status are necessary to better define the association between these disorders and CVD in cats. Future studies will undoubtedly improve understanding of CVD in cats.