Transcranial Doppler sonographic findings in granulomatous meningoencephalitis in small breed dogs

Abstract

Granulomatous meningoencephalitis (GME) is an acute, progressive, and often fatal inflammatory disease of the central nervous system, affecting mainly small and toy dog breeds. A definitive diagnosis of GME can only be achieved through histopathologic examination of samples collected after death. This retrospective study describes transcranial Doppler ultrasonography (TDS) findings in dogs with confirmed clinical histopathology of GME. Eleven dogs were selected for this study. Sonographic findings in B-mode demonstrated diffuse decreased brain parenchyma echogenicity in 9 dogs, ventriculomegaly in 8 dogs, brain atrophy in 4 dogs, and hyperechoic focal lesions in 6 dogs. Color Doppler imaging revealed more obvious vessels of the arterial circle in 10 dogs. Spectral Doppler examination was performed in 10 dogs to detect the 6 major cerebral arteries of interest. The examination showed normal and high resistive index (RI) values in the outlined arteries. The TDS findings were consistent with pathology found on postmortem examination.

Introduction

Granulomatous meningomyeloencephalitis (GME), also known as inflammatory reticulosis, is an acute, non-suppurative, progressive, and often fatal inflammatory disease of the central nervous system (1–7). It is the second most common inflammatory disease of the central nervous system (CNS) in dogs, following canine distemper (1). This disease mainly affects small and toy breed dogs, especially young and middle-aged terriers and poodles (3–7) that have a mean age of 5 y (range: 6 mo to 12 y). The disease affects both sexes, although females show a higher prevalence (4). Lesions are often restricted to the CNS and most commonly occur within the white matter of the cerebrum, cerebellum, caudal brainstem, and cervical spinal cord (7). The etiology of GME remains unknown, although some causes have been proposed and include autoimmune, infectious, and neoplastic diseases.

Three forms of GME have been described based on the site of the lesions. Focal GME has a slowly progressive onset over 3 to 6 mo, with a single space-occupying mass that is commonly located within the brain or spinal cord (8). Disseminated or multifocal GME is an acute and rapidly progressive disease with an onset over 2 to 6 wk, characterized by multiple lesions that cause varied clinical signs (3,7). Ocular GME is an acute form with a static or progressive onset of disease that affects the eyes unilaterally or bilaterally. It is the least common form reported in the literature. It may lead to acute blindness due to optic neuritis; it is occasionally associated with uveitis and more rarely with retinal hemorrhage or detachment (6).

The presumptive diagnosis of GME is based on history, clinical findings, and cerebrospinal fluid (CSF) analysis; however, the differential diagnosis list is long. The clinical signs of a diffuse CNS disease often suggest infection, inflammation or neoplasia, and any benign space-occupying lesion (e.g., neoplasm, inflammatory granuloma, cyst, or infarct) can cause clinical signs of focal CNS disease (1,2). Furthermore, there are many other causes of optic neuritis (6) besides GME. Analysis of the CNS demonstrates mononuclear pleocytosis with increased protein concentration (7,9), which is not a disease-specific finding.

Computed tomography (CT) scan and magnetic resonance imaging (MRI) are useful diagnostic tools for detecting brain injuries (7), but cannot differentiate benign space-occupying masses from neoplastic lesions (1–8). Although surgical brain biopsy is not common in veterinary medicine, it can be a useful alternative procedure for GME diagnosis in animals with focal lesions (4–8).

A definitive diagnosis of GME can only be made postmortem through histopathologic examination (8). Microscopic findings include perivascular cuffs composed of monocytes, macrophages, lymphocytes, and plasma cells (7) and developing granulomas that resemble space-occupying lesions (8).

This disease has a poor prognosis (4), and dogs with the multifocal form have a shorter survival-time (3 to 6 wk following diagnosis) (9) than dogs with the focal form (3 to 6 mo following diagnosis) (4,5).

Transcranial Doppler sonography (TDS) is an alternative method for evaluating the CNS of small breed dogs (10), when CT and MRI are not available. It is rapid, low-cost, and non-invasive compared with CT and MRI, and can be performed without sedation. Transcranial B-mode and Doppler sonography can be performed through an intact skull to assess brain parenchyma (10–14) and cerebral blood flow (15–20), respectively. Small and toy breed dogs are the most suitable (10) for evaluation by these 2 imaging techniques because increased skull bone thickness interferes with ultrasound transmission in large breed dogs.

The purpose of this retrospective study was to describe transcranial Doppler ultrasonography findings in dogs with confirmed clinical histopathology of GME.

Materials and methods

A prospective study of TDS findings was performed on dogs seen between August 2005 and September 2006 at Provet — Private Veterinary Laboratory, Sao Paulo, Brazil. During this period, 116 dogs with clinical signs of CNS disease were subjected to TDS. For this study, we selected 11 dogs with suspected clinical GME that had confirmation of the disease through histopathologic evaluation following euthanasia or spontaneous death.

The TDS examinations were performed by a single individual. The ultrasound equipment used was a GE LOGIQ 3^® (Provet, Sao Paulo, Brazil) with a micro-convex array probe and imaging frequencies ranging from 2 to 5 MHz. Dogs were restrained by the owners and placed into lateral or sternal recumbency without sedation.

A fontanelle window was used, when available, to provide images of the transverse and sagittal planes. A transtemporal window was used in all dogs to provide images of the dorsal and oblique dorsal planes. A suboccipital window was also used in all dogs, through the foramen magnum (accessible when the patient’s head was flexed slightly forward) to provide images of the transverse and sagittal planes.

Initially, gray-scale images were made. The goal of this initial survey was to evaluate the brain parenchyma anatomy by searching for abnormal echogenicity in all the parenchyma, detecting focal lesions and quantifying ventricular width and encephalic mantle thickness. Normal B-mode parameters used in this study have been described previously (10–14).

Color Doppler imaging (CDI) was performed to identify the cerebral arterial circle. Using CDI as a guide, the 6 cerebral arteries of interest (the rostral, middle, and caudal cerebral arteries of both the left and the right hemispheres) were identified. The vessels of the arterial circle were considered normal when observed with minimum adjustment of Doppler color gain settings, without aliasing, as described previously in a technical report (15). Special attention was paid to obtaining the correct insonation angle to maximize the signal, as described previously (15).

Once the vessels were identified, pulsed wave Doppler was initiated, and a spectral tracing with 3 similar sequential waveforms was collected. Measurements were made on a representative spectral waveform to determine peak systolic velocity (PSV), end diastolic velocity (EDV) and resistive index (RI). The relationship between these 3 parameters can be represented as RI = PSV − EDV/PSV. Normal spectral Doppler parameters used in this study have been described previously (15–20).

Results

Of the 11 dogs studied, 7 were female and 4 were male. They ranged from 3 to 15 years of age, with a mean age of 6 y. Six females and 3 males were neutered. There were 5 Yorkshire terriers, 3 miniature poodles, 2 Maltese terriers, and 1 schnauzer. The rostral fontanelle was open in 4 dogs (3 Yorkshire terriers and 1 miniature poodle).

Gray-scale imaging demonstrated that 8 dogs had ventriculomegaly. One of these had asymmetrical lateral dilatation; 7 had symmetrical lateral dilatation (Figure 1), of which 3 also had third ventricle dilation. Using B-mode assessment, decreased brain parenchyma echogenicity was observed in 9 dogs, and brain atrophy with encephalic mantle thickness < 1.25 cm (12,14) was observed in 4 dogs (Table 1). Six dogs had hyper-echoic focal lesions in various parts of the brain. Of these 6 dogs, 1 had a focal lesion in the parietal region of the left cerebral hemisphere and 2 had multifocal lesions. One of the 2 dogs with multifocal lesions had lesions in the temporal and frontoparietal regions of the left cerebral hemisphere, and the other dog had lesions in the cortical region of the right cerebral hemisphere and in the brainstem. Three of the 6 dogs had disseminated lesions: in the right cerebral hemisphere (Figure 2), in the brainstem, and in both the left and right hemispheres. Color Doppler imaging demonstrated conspicuous vessels in the arterial circle that appeared tortuous and more obvious than usual, with minimum adjustments on color gain settings (Figure 3).

Figure 1.

Transtemporal window, dorsal plane. Note the lateral symmetrical ventriculomegaly. LLV — left lateral ventricle, RLV — right lateral ventricle (1, 2 — width of lateral ventricles).

Table 1.

Breed, age, gender, fontanelle, ventricular width and encephalic mantle thickness of the dogs in this study

Patient number	Breed	Age (years)	Gendera	Fontanelle	Ventricular width (cm)		Encephalic mantle thickness (cm)
					RLV	LLV
01	Yorkshire	5	F	Open	1.33	1.36	0.62
02	Maltese terrier	15	F	Closed	0.47	0.48	0.84
03	Yorkshire	3	M	Opened	1.17	0.80	0.93
04	Schnauzer	6	F	Closed	0.29	0.29	2.36
05	Miniature poodle	6	M	Closed	0.25	0.25	2.02
06	Maltese terrier	7	F	Closed	0.38	0.38	2.21
07	Miniature poodle	7	M	Closed	0.62	0.60	1.95
08	Yorkshire	3	F	Closed	0.43	0.48	1.76
09	Miniature poodle	6	F	Open	0.53	0.55	2.26
10	Yorkshire	4	M	Closed	0.25	0.25	1.83
11	Yorkshire	6	F	Open	1.70	1.70	0.52

^aF — female, M — male, RLV — right lateral ventricle, LLV — left lateral ventricle.

Figure 2.

Transtemporal window, oblique dorsal plane. Note between the “x” marks, increased echogenicity areas that demonstrate focal lesions. RCH — right cerebral hemisphere, LCH — left cerebral hemisphere.

Figure 3.

Transtemporal window, oblique dorsal plane. Note that color Doppler imaging was performed to identify the arterial circle. RCA — rostral cerebral artery, MCA — middle cerebral artery, CCA — caudal cerebral artery, LCH — left cerebral hemisphere.

Spectral Doppler imaging was performed on 10 dogs; it was unsuccessful in 1 of the 11 dogs due to the patient’s temperament. Spectral Doppler waveform revealed normal or high RI values (Table 2) that were calculated automatically by the ultrasound equipment. The 6 cerebral arteries of interest were detected in 5 of the dogs; at least 1 artery had high RI values in each dog (Figure 4). In 2 of the dogs, only 1 cerebral artery was undetected. Of those detected, at least 1 cerebral artery had high RI values in each dog. In 3 of the dogs, only 2 arteries were detected. In each of these 3, 1 cerebral artery had high RI values.

Table 2.

Summary of focal lesions and resistive indices in the dogs in the study

Patient	Encephalic parenchyma echogenicity	Focal lesion	Right cerebral hemisphere resistive index (RI)			Left cerebral hemisphere resistive index (RI)
			RCA	MCA	CCA	RCA	MCA	CCA
01	Diffuse hypoechoic	No	0.52	0.55	0.43	0.47	0.55	0.67
02	Diffuse hypoechoic	Hyperechoic multifocal lesions in the temporal and frontoparietal regions of the LCH	0.72	0.87	0,68	0.53	0.58	0.60
03	Diffuse hypoechoic	Hyperechoic focal lesion in the parietal region of the LCH	—	—	—	—	—	—
04	Diffuse hypoechoic	Hyperechoic disseminated lesions in the RCH	0.55	0.81	—	0.44	0.76	0.65
05	Diffuse hypoechoic	Hyperechoic disseminated lesions in the brainstem	0.64	0.52	0.60	0.68	0.55	0.84
06	Diffuse hypoechoic	Hyperechoic multifocal lesions in the cortical region of the RCH and in the brainstem	0.51	0.68	0.64	0.58	0.34	0.45
07	Diffuse hypoechoic	No	0.65	0.60	0.66	0.51	0.51	0.51
08	Normal	No	0.50	0.33	0.50	0.58	—	0.55
09	Diffuse hypoechoic	No	0.53	0.63	—	—	—	—
10	Diffuse hypoechoic	Hyperechoic disseminated lesions in both theLCH and the RCH	0.59	0.64	0.55	0.38	0.54	0.70
11	Normal	No	0.52	0.45	0.61	0.59	0.60	0.55

Color Doppler showed that all 11 dogs had conspicuous vessels in the arterial circle. RCA — rostral cerebral artery, MCA — middle cerebral artery, CCA — caudal cerebral artery, LCH — left cerebral hemisphere, RCH — right cerebral hemisphere.

Figure 4.

Transtemporal window, oblique dorsal plane. Note the normal typical Doppler spectral waveform of the Rostral Cerebral Artery: low resistance and high diastolic velocity pattern. PSV — peak systolic velocity, EDV — end diastolic velocity, RI — resistive index, LCH — left cerebral hemisphere.

Discussion

In this study, GME was most frequently observed in small and toy breed female dogs, with a mean age of 6 y, which is consistent with previous reports (3,12). A comparative assessment was made between pathologic findings and both gray-scale and color Doppler sonographic imaging. Gray-scale imaging showed diffuse decreased brain parenchyma echogenicity in 9 dogs in which histopathologic examination showed congestion and inflammatory changes in the brain tissue. The remaining 2 dogs had normal parenchyma echogenicity although they had a histopathologic diagnosis of GME.

Four dogs (3 Yorkshire terriers and 1 miniature poodle) had open rostral fontanelles, a condition that has a significant relationship with the presence of ventriculomegaly (4,12). Lateral ventriculomegaly, observed in 8 dogs, 7 of which had symmetrical dilatation, was confirmed at necropsy. Previous literature has reported the presence of some degree of hydrocephalus in dogs that have GME (1,8).

Four dogs had a cortical mantle thickness of < 1.25 cm. Three-month-old puppies have normal cortical mantle thickness ranging from 0.65 to 1.25 cm (10,14). However, there are no data on what is considered a normal range for cortical mantle thickness in adult dogs or among different breeds. We compared this finding with the postmortem examination, and brain parenchyma atrophy was confirmed.

Hyperechoic focal lesions observed in 6 dogs, through gray-scale imaging, was compared with the oval-shape granulomas observed in histopathologic examination (1). Through color Doppler imaging, 10 dogs were shown to have conspicuous vessels of the arterial circle. Some of the cortical vessels were visible even with minimum adjustments of color gain. These findings were compared with the diffuse brain congestion and inflammation observed in the histopathologic evaluation of all dogs.

The normal typical Doppler spectral waveform of intracranial arteries demonstrates low resistance and high diastolic velocity pattern (15–20). This index is considered a useful parameter to evaluate cerebrovascular impedance. Spectral Doppler examination was used to observe the 6 major cerebral arteries of interest. Failure to identify some of the vessels of interest through spectral Doppler examination may have been due to mechanical compression (e.g., stenosis or occlusion) caused by the perivascular cuffs observed in pathologic evaluation, or to space-occupying masses found through TDS.

In our study, the dogs showed normal to high RI values. This could be explained by intracranial pressure, which is normal to high in GME patients (4,7).

Although CT and MRI can outline focal lesions (1), the findings are not specific for GME (8). MRI is considered the “gold standard” for diagnostic imaging; by making it possible to visualize the brain tissue without artifacts induced by the cranial bones, it has improved the assessment of anatomic details (8). However, neither ultrasound nor CT scan and MRI can differentiate benign space-occupying lesions from neoplastic ones. MRI and CT play important roles in examining patients with clinical suspicion of GME. However, they are not available in every country, and both are expensive.

To our knowledge, there are no references in the literature to Doppler ultrasonographic findings for GME. Further studies are necessary to verify if the RI of cerebral arteries can be used as a parameter for monitoring the clinical course of GME and to evaluate the effectiveness of therapies for this disease. CVJ