Necrotizing meningoencephalitis in a large mixed-breed dog

• Autores: Chelsie M. Estey
• Localización: JAVMA: Journal of the American Veterinary Medical Association, ISSN-e 0003-1488, Vol. 245, Nº. 11, 2014, págs. 1274-1278
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • Case Description—A 4-year-old 26-kg (57.2-lb) spayed female Staffordshire Bull Terrier mix was evaluated because of a 24-hour history of cluster seizures.

    Clinical Findings—Neurologic examination revealed altered mentation and multifocal intracranial signs; MRI was performed. The MRI findings included multifocal, asymmetric forebrain lesions affecting both the gray and white matter, an area suggestive of focal necrosis, and loss of corticomedullary distinction. A midline shift and caudal transtentorial herniation were noted, suggestive of greater than normal intracranial pressure.

    Treatment and Outcome—Because the dog's clinical signs worsened despite medical treatment and additional evidence of increased intracranial pressure, bilateral craniectomy and durectomy were performed. Histologic evaluation of a brain biopsy specimen revealed bilateral and asymmetric areas of necrosis in the subcortical white matter and adjacent gray matter. At the periphery of the necrotic areas, there was increased expression of glial fibrillary acidic protein and Virchow-Robin spaces were expanded by CD3+ lymphocytes. Results of immunohistochemical analysis of brain tissue were negative for canine distemper virus, Neospora canis, and Toxoplasma gondii. These clinical, imaging, and histopathologic findings were compatible with necrotizing meningoencephalitis. The dog's neurologic status continued to worsen following surgery. Repeated MRI revealed ongoing signs of increased intracranial pressure, despite the bilateral craniectomy. The owners elected euthanasia.

    Clinical Relevance—To the author's knowledge, this is the first report of necrotizing meningoencephalitis in a large mixed-breed dog. Necrotizing meningoencephalitis should be considered as a differential diagnosis in dogs other than small or toy breeds that have signs suggestive of inflammatory disease.

    A 4-year-old 26-kg (57.2-lb) spayed female Staffordshire Bull Terrier mix was evaluated because of a 24-hour history of cluster seizures. The dog had had > 20 seizures in the 12 hours prior. The dog's vaccine status was current but it had not been vaccinated recently and had been otherwise healthy.

    At the initial examination, the dog was nonambulatory and alternated between obtundation and stupor. Heart rate, respiratory rate, and rectal temperature were within reference ranges. Neurologic examination during periods of obtundation revealed head pressing, leaning to the right, nonambulatory tetraparesis, unresponsive miotic pupils and absent menace responses in both eyes, decreased paw replacement for all limbs, and an absent hopping response on both the right thoracic and pelvic limbs. Other neurologic examination findings were unremarkable. The neuroanatomic localization was multifocal (forebrain and brainstem). Initial clinicopathologic findings included mild hypokalemia (3.2 mEq/L; reference range, 3.8 to 5.4 mEq/L); mildly low creatinine concentration (0.5 mg/dL; reference range, 0.6 to 1.4 mg/dL); mild hypercholesterolemia (343 mg/dL; reference range, 138 to 332 mg/dL); mild normocytic, normochromic, nonregenerative anemia; and mild leukocytosis (16 × 103 leukocytes/μL; reference range, 5.7 × 103 leukocytes/μL to 14.2 × 103 leukocytes/μL). During the initial examination, the dog had a seizure. A dose of levetiracetama (20 mg/kg [9.1 mg/lb], IV) was administered, along with a dose of mannitolb (0.5 g/kg [0.23 g/lb], IV).

    Because of the severity of the dog's neurologic condition and a change in the dog's mentation from stuporous to comatose, MRIc of the brain was performed. Poorly marginated areas of hyperintensity (relative to normal brain tissue) were noted diffusely throughout the entire cerebrum on T2-weighted (repetition time, 4,200 milliseconds; echo time, 90 milliseconds; slice thickness, 3 mm) and T2 fluid-attenuated inversion recovery images (repetition time, 8,100 milliseconds; echo time, 90 milliseconds; slice thickness, 3 mm). Periventricular hyperintensity was also present at the level of the lateral ventricles. These areas were isointense, compared with gray matter, on T1-weighted images (repetition time, 650 milliseconds; echo time, 10 milliseconds; slice thickness, 3 mm) obtained prior to contrast agent administration, with no enhancement following IV administration of gadolinium. Loss of normal surface topography (ie, inapparent sulci and gyri), a mild rightward midline shift, and caudal transtentorial herniation of the cerebrum, with secondary compression of the fourth ventricle and rostral aspect of the cerebellum (Figure 1), together suggested increased intracranial pressure. Loss of distinction between the gray and white matter was also evident diffusely within the cerebral hemispheres. A focal lesion was detected within the white matter of the right occipital lobe, dorsal to the lateral ventricle; the lesion appeared hyperintense relative to white and gray matter on T2-weighted images, hypointense relative to gray matter on T1-weighted images, and had a hyperintense rim and isointense center (relative to gray and white matter) on T2 fluid-attenuated inversion recovery images. Neither contrast enhancement nor mass-effect was evident in association with this lesion. This area was considered most consistent with an area of necrosis. A sample of CSF was not collected because of presumed high intracranial pressure.

    View larger version(31K) Figure 1— Transverse (A) and sagittal (B) T2-weighted MRI images of the brain of a 4-year-old spayed female Staffordshire Bull Terrier mix with a 24-hour history of cluster seizures. In both views, notice the focal hyperintense lesion within the right occipital lobe (arrows) that is most consistent with an area of necrosis. In panel B, diffusely decreased corticomedullary distinction (arrowhead) and caudal transtentorial herniation (asterisk) are evident.

    Differential diagnoses for the dog's condition primarily included infectious or meningoencephalitis of unknown cause (GME, NME, or NLE) and neoplasia. The dog was treated overnight with doxycyclined (5 mg/kg [2.3 mg/lb], IV, q 12 h), clindamycine (10 mg/kg [4.5 mg/lb], IV, q 12 h), famotidinef (0.5 mg/kg [0.23 mg/lb], IV, q 12 h), dexamethasone sodium phosphateg (0.08 mg/kg [0.036 mg/lb], IV, q 12 h), and levetiracetam (20 mg/kg [9.1 mg/lb], IV, q 8 h). Serial neurologic examinations initially revealed nonresponsive pupils followed by hippus. Blood pressure and heart rate fluctuated erratically. Results of clinicopathologic analyses performed the following day were relatively unchanged from those at admission. The dog's condition continued to worsen, and neurologic examination revealed mydriasis in both eyes, comatose mentation, and signs consistent with a Cushing's response (ie, concurrent hypertension and bradycardia). Functional brainstem reflexes, such as pupillary light reflexes and palpebral and dazzle reflexes in both eyes were taken to signify residual brainstem function. As such, brainstem auditory response testing was not performed.

    Attempts to decrease intracranial pressure at that time through mannitol and corticosteroid administration were not successful. On discussion of clinical status and prognosis with the dog's owners, and recognizing that the dog's more urgent problem was of further brain herniation due to refractory increases in intracranial pressure, a bilateral decompressive craniectomy was offered as an alternative means of reducing intracranial pressure and of allowing time for medical management to potentially have a positive effect. The owners were made aware of the risk of the potential lack of sufficient brain function after surgery to allow for an acceptable quality of life but elected to proceed despite this concern. A bilateral craniectomy (transparietal approach) and durectomy were performed to obtain brain biopsy specimens and reduce intracranial pressure. On removal of the dura, the cerebral cortex protruded from the craniectomy sites bilaterally. A focal 1-cm area of purple discoloration was present on the surface of the right occipital lobe and extended into the deeper white matter tracts, corresponding with the lesion seen on MRI. This site was biopsied, as was an irregular linear region rostral to this lesion. Both impression smears and formalin-fixed surgical biopsy specimens of these areas were submitted for analysis. Impression smears were examined microscopically and findings were consistent with mild lymphocytic and macrophagic inflammation.

    Because of lack of improvement after craniectomy and a recurrence of seizures, MRI was repeated. The cerebrum remained diffusely enlarged and was seen protruding from the margins of the craniectomy sites bilaterally. The MRI findings remained otherwise unchanged from those of the previous MRI examination (Figures 2 and 3) aside from mild improvement in the degree of transtentorial herniation after surgery. Given the continual decline in the dog's condition following surgery and MRI evidence of ongoing mass effect and herniation despite the bilateral craniectomy, the owners elected euthanasia. Although the owners declined a full necropsy, they did permit examination of the entire brain. Both the biopsy specimens collected during craniectomy and the brain were formalin fixed, embedded in paraffin, and stained with H&E stain for histologic examination.

    View larger version(30K) Figure 2— Midsagittal, T2-weighted MRI images of the brain of the dog in Figure 1 before (A) and after (B) craniectomy and durectomy. After surgery, mild improvement in the degree of transtentorial herniation is visible.

    View larger version(32K) Figure 3— Transverse fluid-attenuated inversion recovery MRI images of the brain of the dog in Figure 1 at the level of the midbrain before (A) and after (B) craniectomy and durectomy. In panel A, notice the hyperintense lesion in the area of the right occipital lobe (arrow) and periventricular hyperintensity. In panel B, herniation of the left cerebral hemisphere through the craniectomy site is evident (arrowhead).

    Histologic examination of the brain biopsy specimens revealed extensive areas of necrosis in the subcortical white matter and deep layers of the gray matter in the right cerebral hemisphere, which were infiltrated by large numbers of gitter cells and astrocytes (astrogliosis; Figure 4). Within and adjacent to the areas of necrosis, lymphocytes expanded the Virchow-Robin spaces (perivascular cuffs). The overlying leptomeninges were variably expanded by mild numbers of neutrophils and fewer histiocytes, lymphocytes, and plasma cells (Figure 5). The contralateral cerebral hemisphere had similar, yet less severe, lesions targeting the leptomeninges and the cortical gray and white matter. There was increased glial fibrillary acidic protein staining surrounding the areas of necrosis, and approximately 50% of lymphocytes cuffing cerebral vessels were immunoreactive for CD3. Bacteria and fungal organisms were not detected in Gram- or periodic acid-Schiff–stained sections, respectively. Results of immunohistochemical analysis of brain tissue were negative for canine distemper virus, Neospora caninum, and Toxoplasma gondii. The distribution and nature of the necrotizing lesions were most compatible with those described in Pugs, Malteses, and Chihuahuas with NME.1–4 View larger version(84K) Figure 4— Photomicrograph of a section of the right cerebral hemisphere of the dog in Figure 1. The subcortical white matter contains extensive areas of malacia and gliosis, resulting in blurring of the gray and white matter junction (long arrow). There is moderate nonsuppurative inflammation surrounding vessels and expanding the leptomeninges (short arrows). H&E stain; bar = 1 mm.

    View larger version(95K) Figure 5— Photomicrograph of another section of the same cerebral hemisphere as in Figure 4. The Virchow-Robin spaces are expanded by large numbers of lymphocytes. H&E stain; bar = 200 μm. Inset—Approximately 50% of lymphocytes are immunoreactive for CD3.