IMAGING DIAGNOSIS—MAGNETIC RESONANCE IMAGING PULSATILITY ARTIFACT IN THE CANINE CERVICAL SPINE

Pulsatile venous flow in the internal vertebral venous plexus of the cervical spine can lead to vertical, linear T2‐hyperintensities in the spinal cord at the cranial aspect of C3 and C4 in transverse T2‐weighted images in large breed dogs that are not accompanied by ghosting. The artifact is more conspicuous in pre‐ and postcontrast transverse T1‐weighted images and is accompanied by ghosting in that sequence, typical of a pulsatility artifact. A flow‐related artifact was confirmed as the cause for this appearance by noting its absence after either exchange of phase and frequency encoding direction or by flow compensation. Care should be exercised to avoid misdiagnosing this pulsatility artifact seen in transverse T2‐weighted images of the midcervical spine in large dogs as an intramedullary lesion when T1‐images or phase‐swap images are not available to confirm its artifactual origin.     

APPEARANCE OF THE CANINE MENINGES IN SUBTRACTION MAGNETIC RESONANCE IMAGES

The canine meninges are not visible as discrete structures in noncontrast magnetic resonance (MR) images, and are incompletely visualized in T1‐weighted, postgadolinium images, reportedly appearing as short, thin curvilinear segments with minimal enhancement. Subtraction imaging facilitates detection of enhancement of tissues, hence may increase the conspicuity of meninges. The aim of the present study was to describe qualitatively the appearance of canine meninges in subtraction MR images obtained using a dynamic technique. Images were reviewed of 10 consecutive dogs that had dynamic pre‐ and postgadolinium T1W imaging of the brain that was interpreted as normal, and had normal cerebrospinal fluid. Image‐anatomic correlation was facilitated by dissection and histologic examination of two canine cadavers. Meningeal enhancement was relatively inconspicuous in postgadolinium T1‐weighted images, but was clearly visible in subtraction images of all dogs. Enhancement was visible as faint, small‐rounded foci compatible with vessels seen end on within the sulci, a series of larger rounded foci compatible with vessels of variable caliber on the dorsal aspect of the cerebral cortex, and a continuous thin zone of moderate enhancement around the brain. Superimposition of color‐encoded subtraction images on pregadolinium T1‐ and T2‐weighted images facilitated localization of the origin of enhancement, which appeared to be predominantly dural, with relatively few leptomeningeal structures visible. Dynamic subtraction MR imaging should be considered for inclusion in clinical brain MR protocols because of the possibility that its use may increase sensitivity for lesions affecting the meninges.     

MAGNETIC RESONANCE IMAGING OF THE CANINE ABDOMEN: EFFECT OF PULSE SEQUENCE ON DIAGNOSTIC QUALITY

Motion artifact is an important limiting factor for abdominal magnetic resonance imaging (MRI) in veterinary patients. The purpose of this study was to determine the effects of pulse sequence on abdominal MRI diagnostic quality in dogs. Ten normal dogs were each scanned using 16 MRI pulse sequences. Sequences included breath‐holding sequences, respiratory navigation sequences, and traditional spin‐echo sequences. Four observers independently scored diagnostic quality for each sequence based on the appearance of specific organs, overall diagnostic quality, and degree of artifactual interference. Signal‐to‐noise ratio and contrast‐to‐noise ratio were also calculated for each sequence. The sequence with the highest overall mean diagnostic quality score was the dorsal T2 turbo spin echo (TSE) with fat saturation and breath‐holding. The sequence with the lowest mean diagnostic quality score was the dorsal T2 fast spin echo. The sequence with the highest signal‐to‐noise ratio for all evaluated organs was the sagittal T1 spin echo. Signal‐to‐noise and contrast‐to‐noise ratios did not correlate with subjective assessment of overall diagnostic quality for the majority of the sequences evaluated (P < 0.05). The three sequences considered to have the highest diagnostic quality for the cranial abdomen were the dorsal T2 TSE with fat saturation and breath‐hold, transverse T1 turbo fast low‐angle shot gradient echo with breath‐hold, and dorsal T2 half‐Fourier acquisition single shot TSE with respiratory navigation. These sequences had short acquisition times, yielded studies of similar diagnostic quality, provided complementary information, and are therefore recommended for routine canine abdominal MRI protocols.     

MAGNETIC RESONANCE IMAGING FINDINGS IN SPINAL CORD INFARCTION IN THREE SMALL BREED DOGS

Fibrocartilaginous embolization (FCE) of the spinal cord is a common disease in large breed dogs. There are only a few reports about this entity in small breed dogs and it has never been reported in chondrodystrophic breed. For definitive diagnosis histopathologic examination is necessary. Magnetic resonance imaging (MRI) as a potential diagnostic tool for intravitam diagnosis of FCE has been mentioned before, but results have not been reported so far. This report describes the neurological findings and MRI results in three small breed dogs, including a Pekingese dog, with FCE of the spinal cord. The disease was suspected in two animals based upon clinical and MRI-appearance and confirmed in the third by histopathological examination. In all three cases, similar focal intramedullary lesions, consisting of hyperintensive signals on T2-weighted images, were detected. Based on these findings, high-field MRI may be used as an antemortem tool for the diagnosis of FCE. It is also shown that FCE can occur in chondrodystrophic dogs.     

IMAGING DIAGNOSIS—MAGNETIC RESONANCE IMAGING FEATURES OF A MULTIFOCAL OLIGODENDROGLIOMA IN THE SPINAL CORD AND BRAIN OF A DOG

An 8‐year‐old neutered male Toy Poodle was presented with chronic, progressive tetraparesis, and possible seizures. Magnetic resonance images demonstrated an extensive, T1 and T2 hyperintense contrast enhancing mass in the cervical spinal cord. Three nodules were present on the surface of the thalamus, with enhancement most evident on delayed images. A diagnosis of high‐grade oligodendroglioma was confirmed with postmortem histopathology and immunohistochemical labeling. Oligodendroglioma should be considered as a differential for T1 hyperintense intraaxial or intramedullary lesions with contrast enhancement. If enhancement is not visualized on postcontrast images, delayed images may be beneficial.     

ULTRASONOGRAPHIC ANATOMY OF THE NORMAL CANINE SPINAL CORD AND CORRELATION WITH HISTOPATHOLOGY AFTER INDUCED SPINAL CORD TRAUMA

Prior to trauma, intraoperative ultrasound of the spinal canal in 31 normal dogs was performed through a hemilaminectomy in the left pedicle of L2. A ventral compressive model of spinal cord injury was performed as part of a clinical drug trial. Maximum ultrasonographic spinal cord diameter ranged from 4.9–7.2 mm (5.7 × 0.6). Significant positive correlation (p = 0.023, r = 0.49) was found between age and spinal cord diameter. The dura mater was a separate, well-defined, echogenic horizontal line in 28 (90%) dogs, dorsally, and in 29 (94%) dogs, ventrally. Cerebrospinal fluid was anechoic. Eighteen (58%) dogs had a well-defined anechoic dorsal subarachnoid space, whereas 22 (71%) had a well-defined ventral space. Pia mater was thin but strongly echogenic and covered spinal cord. Central canal was a double hyperechoic line in 17 (55%) dogs and a single-line in 14 (45%) dogs. A difference in the ultrasonographic appearance between gray and white matter was not seen. Epidural fat and connective tissue was a lobular echogenic material in the ventral epidural space. The periosteal-vertebral body interface was seen as a bright curvilinear echo with distal acoustic shadowing. Spinal cord parenchyma could be classified subjectively into four groups based upon ultrasonographic appearance. Spinal cord parenchyma had a uniform hypoechogenicity in 8 (27%) dogs (Group 1), subtle low level echoes in 7 (23%) dogs (Group 2), multiple clusters of defined echogenic foci in 12 (37%) dogs (Group 3), and multiple sharply-defined linear echoes in 4 (13%) dogs (Group 4). There was a significant relationship between pre-trauma ultrasonographic appearance of the spinal cord and histopathology 21 days after trauma. One (13%) dog in Group 1, 4 (57%) dogs in Group 2,10 (91%) dogs in Group 3, and 3 (75%) dogs in Group 4 had malacia on histological evaluation. Therefore, dogs with echogenic spinal cords or linear echoes within cord parenchyma were significantly more likely to develop malacia rather than Wallerian degeneration after induced spinal cord trauma (p = 0.002). Spinal cord echogenicity may indicate vascularity in a segment of spinal cord and might be prognostic following spinal cord trauma. No complications were found related to intraoperative ultrasound. Hematoma or fibrous tissue formation appeared to impede percutaneous ultrasound of the spinal cord in dogs re-evaluated forty-eight hours and one week after surgery.     

COMPARISON OF CONVENTIONAL SPIN-ECHO AND FAST SPIN-ECHO MAGNETIC RESONANCE IMAGING IN THE CANINE BRAIN

T2-weighted fast spin echo and conventional spin echo are two magnetic resonance (MR) pulse sequences used to image the brain. Given the same scan parameters the resolution of fast spin-echo images will be inferior to that of conventional spin-echo images. However, fast spin-echo images can be acquired in a shorter time allowing scan parameters to be optimized for increased resolution without increasing the time to an unacceptable level. MR imaging of the brain of 54 dogs, suspected of having parenchymal brain abnormalities was performed using a 1.5 T scanner. Acquisition time ranged from 4 min 24 s to 7 min 16 s (average = 5 min 15 s) for fast spin-echo scans and from 6 min 32 s to 11 min 26s (average = 7 min 55s) for conventional spin-echo scans. All reviewers consistently rated the resolution of fast spin-echo images higher than the conventional spin-echo images (P = 0.000). The potential disadvantages of fast spin-echo acquisitions (motion artifacts, blurring, and increased hyperintensity of fat) did not affect the resolution of the images. Fast spin echo offers increased resolution in a comparable time to conventional spin echo by increased number of excitations and finer matrix size, thus improving the signal-to-noise ratio and spatial resolution, respectively.     

MAGNETIC RESONANCE IMAGING FEATURES OF TUMORS OF THE SPINE AND SPINAL CORD IN DOGS

Twenty-one dogs with confirmed tumors of the spinal cord or paraspinal tissues were imaged with magnetic resonance (MR) imaging. Anatomical location, location in relation to the dura and the medulla (spinal cord), and bone infiltration were assessed on the MR images and compared to findings at surgery or necropsy. Localization of tumors in the intradural-extramedullary compartment was not always possible. Bone infiltration was correctly assessed in all but one dog, and the anatomical locations involved were accurately determined in all dogs. Sagittal T2-weighted images were helpful to determine the anatomical location. Transverse T1-weighted images pre and post Gd-DTPA administration were helpful for additional localization and definition of tumor extension.     

THE USE OF MAGNETIC RESONANCE IMAGING IN EVALUATING HORSES WITH SPINAL ATAXIA

To determine the accuracy of magnetic resonance imaging for diagnosing cervical stenotic myelopathy in horses, 39 horses with spinal ataxia and 20 control horses underwent clinical and neurologic examinations, cervical radiographs, euthanasia, magnetic resonance (MR) imaging of the cervical spine and necropsy. Twenty‐four horses were diagnosed with cervical stenotic myelopathy, 5 with cervical vertebral stenosis, 7 with idiopathic ataxia, 3 horses had other causes of ataxia, and 20 were controls. The MR images were assessed for spinal cord intensity changes, presence of spinal cord compression, spinal cord compression direction, shape of spinal cord, and the presence of synovial cysts, joint mice, and degenerative joint disease. The height, width, and area of the spinal cord, dural tube and vertebral canal were measured. The identification of spinal cord compression on MR images was significantly different in horses with cervical stenotic myelopathy (P < 0.02), but in the cervical stenotic myelopathy group the identification of spinal cord compression on MR images had poor to slight agreement with histopathologic evidence of compression (κ = 0.05). Horses with cervical stenotic myelopathy were more likely to have a T2 hyperintensity in the spinal cord (P < 0.05). Horses with cervical stenotic myelopathy or cervical vertebral stenosis were more likely to have degenerative joint disease than control horses or horses with other or idiopathic ataxia.     

COMPARISON OF FAST SPIN-ECHO AND CONVENTIONAL SPIN-ECHO MAGNETIC RESONANCE SPINAL IMAGING TECHNIQUES IN FOUR NORMAL DOGS

Various magnetic resonance (MR) imaging techniques have been used to assess lumbar spinal abnormalities in people. Four, young adult, clinically normal dogs were used to compare images of the spinal cord acquired using conventional spin-echo and rapid acquisition relaxation-enhanced (RARE), commonly called fast spin-echo (FSE), magnetic resonance imaging techniques. Lateral myelograms were made as an anatomic control. The T2-weighted FSE technique was characterized by better image quality than the T2-weighted conventional spin-echo technique. The short acquisition time with the FSE technique allowed increases in the matrix size and number of excitations, thus improving resolution and signal-to-noise ratio. In canine lumbar spinal MR imaging, use of a FSE technique is recommended to reduce the overall time for imaging and to improve image quality.     

POSTOPERATIVE IRRADIATION OF SPINAL CORD TUMORS IN 9 DOGS

Between 1985 and 1993, nine dogs with spinal cord tumors were treated postoperatively with cobaltradiation at North Carolina State University-Veterinary Teaching Hospital. Total doses ranged between 33.3–48.0 Gy given in 10–12 fractions of 3–4 Gy over a four week period. Five dogs were euthanized due to recurrence of the tumor or neurologic signs and two dogs were euthanized due tounrelated problems. Two dogs were alive but lost to follow-up at 12 and 25 months. Survival time ranged from 6.5–70.0 months. Median survival time (95% confidence interval) was 17 (12–70) months. Results of this study suggest decompressive surgery followed by irradiation can be an effective treatment for dogs with spinal cord tumors.     

MAGNETIC RESONANCE IMAGING FEATURES OF CERVICAL SPINAL CORD MENINGIOMAS

The records of four dogs with cervical spinal cord meningiomas were retrospectively reviewed. Signalment, history, laboratory findings, neurological examination, and histopathological findings were evaluated. Magnetic resonance imaging (MRI) was performed using a 1.0-T superconducting magnet and T2-weighted (W) and noncontrast and postcontrast T1-W spin echo pulse sequences. Meningiomas were located at the level of the second, third, and fifth cervical vertebrae and the C2-3 intervertebral space. All meningiomas appeared as focal masses that were hyperintense to the spinal cord on T2-W images and iso- to hypointense on the T1-W images. They could be identified as intradural and extramedullary in origin based on a broad-based dural margin seen on at least one of the imaging planes and a gradual expansion of the subarachnoid space cranial and caudal to the mass, best noted on the transverse and dorsal plane images. On dorsal plane T2-W images in three dogs, expansion of the subarachnoid space adjacent to the mass appeared similar to the myelographic "golf tee" sign. All meningiomas exhibited moderate, well-defined contrast enhancement with dural tails seen in three of the four dogs. One dog had extension into the intervertebral foramen along the nerve and ipsilateral atrophy of the muscles of the neck. By differentiating the meningiomas from intramedullary tumors and by clearly depicting the extent of the masses, MRI provided valuable information about treatment options and prognosis.     

THE USE OF DIFFUSION TENSOR IMAGING TO EVALUATE THE SPINAL CORD IN NORMAL AND ABNORMAL DOGS

Diffusion tensor imaging (DTI) is a specialized magnetic resonance sequence to determine the direction of water molecule motion. Our hypothesis was that information derived from DTI will be significantly different in dogs with a spinal cord lesion compared with a normal dog. Eleven normal dogs and six dogs with a spinal cord lesions were imaged. DTI was performed along with standard T1‐ and T2‐weighted sequences in transverse and sagittal planes. Fractional anisotrophy and apparent diffusion coefficient (ADC) were obtained using regions of interests centered on the cranial aspect, middle cranial, middle caudal, and caudal aspects of the spinal cord. In normal dogs, the DTI sequence was characterized by normal fiber tracking with no statistical difference between the four sections of spinal cord (P>0.05). In the dogs with a spinal cord lesion, there was a significant difference in fractional anisotropy between the two groups (P=0.0003) and the ADC analysis statistical significance (P=0.048) at the caudal most site. Based on these findings, DTI is a potentially useful method to evaluate the spinal cord in dogs.     

DIFFUSION‐WEIGHTED MAGNETIC RESONANCE IMAGING OF THE NORMAL CANINE BRAIN

Diffusion‐weighted imaging (DWI) MRI has been primarily reported as a method for diagnosing cerebrovascular disease in veterinary patients. In humans, clinical applications for diffusion‐weighted MRI have also included epilepsy, Alzheimer's, and Creutzfeld–Jakob disease. Before these applications can be developed in veterinary patients, more data on brain diffusion characteristics are needed. Therefore, the aim of this study was to evaluate the distribution of diffusion in the normal canine brain. Magnetic resonance imaging of the brain was performed in ten, clinically normal, purpose‐bred beagle dogs. On apparent diffusion coefficient maps, regions of interest were drawn around the caudate nucleus, thalamus, piriform lobe, hippocampus, semioval center, and cerebral cortex. Statistically significant differences in mean apparent diffusion coefficient were found for the internal capsule, hippocampus, and thalamus. The highest apparent diffusion coefficient (1044.29 ± 165.21 μm²/s (mean ± SD (standard deviation)) was detected in the hippocampus. The lowest apparent diffusion coefficient was measured in the semioval center (721.39 ± 126.28 μm²/s (mean ± SD)). Significant differences in mean apparent diffusion coefficients of the caudate nucleus, thalamus, and piriform lobe were found by comparing right and left sides. Differences between brain regions may occur due to differences in myelination, neural density, or fiber orientation. The reason for the differences between right and left sides remains unclear. Data from the current study provide background for further studies of diffusion changes in dogs with brain disease.     

USE OF INTRAOPERATIVE ULTRASONOGRAPHY IN CANINE SPINAL CORD LESIONS

The purpose of this retrospective study was to describe the intraoperative appearance of various spinal cord conditions, and to investigate how intraoperative ultrasonography assisted in modification of surgical and postoperative treatment plans. Intraoperative ultrasonography (B-mode, and power Doppler mode) was used in 25 dogs undergoing spinal surgery. The neurologic conditions included cervical spondylomyelopathy, intervertebral disc (IVD) protrusion, IVD extrusion, spinal tumors, nerve sheath mass, granulomatous myelitis, and discospondylitis. All of these diagnoses were supported by histopathologic and/or cytologic evaluation. It was possible to visualize the spinal cord and the abnormal spinal tissue in all of the patients. Power Doppler imaging allowed assessment of the spinal cord microcirculation, and assisted in judgment of the degree of decompression. Ultrasound imaging directly impacted the surgical and the medical treatment plans in four patients. Owing to the intraoperative imaging, two hemilaminectomies were extended cranially and caudally, and additional disc spaces were fenestrated, one hemilaminectomy site was extended dorsally to retrieve the disc material from the opposite side, and one intramedullary cervical spinal cord lesion was discovered, aspirated, and consequently diagnosed as granulomatous inflammation, which altered the long-term medication protocol in that dog. This study suggests that intraoperative sonographic spinal cord imaging is a useful and viable technique.     

MAGNETIC RESONANCE IMAGING IN THE DIAGNOSIS OF CANINE INFLAMMATORY MYOPATHIES IN THREE DOGS

In humans affected with inflammatory myopathies, regions of altered signal intensity are found on magnetic resonance (MR) images of affected muscles. Although electromyography (EMG) is more practical for muscle disease evaluation, and a muscle biopsy is the only manner in which a definitive diagnosis can be made, MR imaging has proven useful if a specific anatomic localization is difficult to achieve. Three dogs with focal inflammatory myopathy diagnosed with the assistance of MR imaging are discussed and the findings are compared with those found in humans. MR images of the affected muscles in each dog were characterized by diffuse and poorly marginated abnormal signal on T1- and T2-weighted images. Marked enhancement was noted in these muscles after contrast medium administration. An inflammatory myopathy was confirmed histologically in all three dogs. A good association existed between the MR images and muscle inflammation identified histopathologically. MR imaging may be a useful adjunctive procedure for canine inflammatory myopathies.     

MAGNETIC RESONANCE IMAGING APPEARANCE OF SUSPECTED ISCHEMIC MYELOPATHY IN DOGS

Ischemia and infarction of the spinal cord is a known cause of acute spinal injury in dogs. Currently, the diagnosis of spinal cord infarction in small animals is based on history, clinical signs, and the exclusion of other differentials with radiography and myelography. It is a diagnosis only confirmed through necropsy examination of the spinal cord. The aim of this paper is to describe the Magnetic resonance imaging (MRI) findings of the spinal cord of dogs with suspected spinal cord infarcts to utilize this technology for antemortem support of this diagnosis. This retrospective study evaluated the spinal MR examinations of 11 dogs with acute onset of asymmetric nonpainful myelopathies. All patients except one (imaged at 2 months) were imaged within 1 week of clinical signs and managed conservatively with minimal medical and no surgical intervention. They were followed clinically for a minimum of 4 months after discharge. MR findings in all dogs were characterized by focal, intramedullary, hyperintense lesions on T2-weighted images with variable contrast enhancement similar to what is reported in humans. Though it could not be used to diagnose spinal cord infarction definitively, MRI was useful in excluding extramedullary spinal lesions and supporting intramedullary infarction as a cause of the acute neurologic signs. Together with the history and clinical examination findings, MRI is supportive of a diagnosis of spinal cord infarction.     

COLOR DOPPLER AND DOPPLER SPECTRAL ANALYSIS IN THE SPINAL CORD OF NORMAL DOGS

Doppler ultrasonography of the spinal cord was performed in 34 normal, anesthetized dogs following hemilaminectomy. This study was part of an investigation to evaluate the efficacy of a 21-aminosteroid compound and high dose methylprednisolone for the treatment of spinal cord trauma. Grey-scale images of the canine spinal cord were similar to those described for the spinal cord of people. Doppler waveforms of intraparenchymal spinal arteries exhibited high end diastolic blood flow velocities, indicating low resistance to flow. Doppler values (mean ± SD) for arteries immediately ventrolateral to the central canal were: Peak Systolic Velocity = 5.78 ± 2.5 cm/sec, Minimum Diastolic Velocity = 3.5 ± 1.62 cm/sec, Mean Velocity = 4.45 ± 1.96 cm/sec, Minimum Diastolic Velocity = 3.5 ± 1.62 cm/sec, Mean Velocity = 4.45 ± 1.96 cm/sec, Systolic/Diastolic ratio = 169 ± 0.19, Pulsatility Index = 0.53 ± 0.09, and Resistance Index = 0.4 ± 0.06.     

MAGNETIC RESONANCE IMAGING OF THE CERVICAL SPINE IN 27 DOGS

The cervical spine of 27 dogs with cervical pain or cervical myelopathy was evaluated using magnetic resonance imaging (MRI). Spin echo T1, T2, and post-contrast T1 weighted imaging sequences were obtained with a 0.5 Tesla magnet in 5 dogs and a 1.5 Tesla magnet in the remaining 22 dogs. MRI provided for visualization of the entire cervical spine including the vertebral bodies, intervertebral discs, vertebral canal, and spinal cord. Disorders noted included intervertebral disc degeneration and/or protrusion (12 dogs), intradural extramedullary mass lesions (3 dogs), intradural and extradural nerve root tumors (3 dogs), hydromyelia/syringomyelia (1 dog), intramedullary ring enhancing lesions (1 dog), extradural synovial cysts (1 dog), and extradural compressive lesions (3 dogs). The MRI findings were consistent with surgical findings in 18 dogs that underwent surgery. Magnetic resonance imaging provided a safe, useful non-invasive method of evaluating the cervical spinal cord.