History
A 9-year-old spayed female Greyhound that was nonambulatory was examined because of painful progressive paraparesis of 1 week's duration. No precipitating cause had been identified by the owner. Findings on initial CBC and serum biochemical analysis were unremarkable.
Physical examination findings were unremarkable. Neurologic assessment revealed paraparesis with absent conscious proprioception in both pelvic limbs and normal spinal reflexes. The cutaneous trunci reflex was absent caudal to L4. Signs of diffuse pain were evident on palpation of the cranial lumbar region. The lesion was localized to the T3-L3 spinal cord segment. Survey radiography of the lumbar portion of the vertebral column revealed mineralization of several intervertebral disks and narrowing of the intervertebral disk space at L1-2. Magnetic resonance imaging (MRI) of the lumbar vertebral column was performed under general anesthesia (Figure 1).
Spin echo T1-weighted (T1-W; A), spin echo T2-weighted (T2-W; B), and short T1 inversion-recovery (STIR; C) sagittal magnetic resonance images of the lumbar vertebral column of a 9-year-old spayed female Greyhound that was nonambulatory and was examined because of painful progressive paraparesis of 1 week's duration. Transverse T1-W (D), T2-W (E), and T2* gradient echo (F) images were taken at the level depicted by the vertical line on image A. L = Left.
Citation: Journal of the American Veterinary Medical Association 235, 7; 10.2460/javma.235.7.823
Determine whether additional imaging studies are required, or make your diagnosis from Figure 1—then turn the page →
Diagnostic Imaging Findings and Interpretation
An extradural lesion at the level of the L1-2 intervertebral disk space with left lateral and ventral compression of the spinal cord that extends cranially and caudally is evident (Figure 2). Three regions are observed. The most cranial part is a well-defined structure above the L1 vertebral body that is hyperintense, compared with the spinal cord, in spin echo T1-weighted (T1-W) and T2-weighted (T2-W) sequences. Short T1 inversion-recovery (STIR) sequences do not suppress the area of bright signal intensity, which indicates that the lesion has no fat component; hence, hemorrhage is to be considered. However, on T2* gradient echo (GE) sequences, the region has a hyperintense signal, compared with that of the spinal cord, ruling out free hemosiderin. Caudal to this region, the extradural compression is of mixed signal intensity, which is predominantly isointense to the spinal cord in T1-W sequences and hyperintense to the spinal cord in T2-W sequences. In this area, T2* GE images reveal regions of signal void, which indicate hemorrhage. In the caudal-most aspect of the lesion, there is an area with bright signal intensity in T1-W and T2-W sequences, which is suppressed by STIR and suggests epidural fat. There is also a diffuse hyperintensity in T2-W sequences within the spinal cord cranial and caudal to the extradural compression. These intramedullary changes are compatible with spinal cord edema. The T1-W images obtained after the administration of gadopentetate dimeglumine did not reveal abnormal uptake.
Same magnetic resonance images as in Figure 1. Signal intensity in all sequences is compared with the normal signal intensity of the spinal cord. A, B, and C—Sagittal images reveal a complex extradural lesion compressing the spinal cord at the level of L1-L2. There are 3 areas within this extradural lesion. The most cranial compartment (solid arrows) has a high signal intensity in T1-W images and is hyperintense in T2-W and STIR sequences. Caudal to this (outlined arrows), the extradural lesion has a mixed signal intensity, being predominantly hypointense in T1-W and hyperintense in T2-W images. Notice the region that is hyperintense in both T1-W and T2-W images and is suppressed by STIR (arrowhead). Notice the high signal intensity in T2-W images within the spinal cord (asterisk). This extends cranial and caudal to the extradural lesion. D, E, and F—Transverse images reveal the spinal cord (outlined arrows) and extradural lesion (solid arrows). These images confirm the well-defined extradural structure (solid arrows) to be hyperintense in T1-W (D) and T2-W sequences (E) with no areas of signal void on the T2* gradient echo image (F). See Figure 1 for remainder of key.
Citation: Journal of the American Veterinary Medical Association 235, 7; 10.2460/javma.235.7.823
The MRI diagnosis is extradural spinal cord compression, which, based on the narrowing of the intervertebral disk space at L1-2, the extradural lesion just dorsal to this disk space, and lack of uptake, is compatible with intervertebral disk extrusion at L1-L2. The complex appearance of an unusual hyperintense signal in T1-W images of the extradural lesion is suggestive of disk material, accompanied by extradural hemorrhage, epidural fat, and spinal cord edema.
Comments
This dog had an unusual MRI appearance of Hansen Type I intervertebral disk disease, which involves mineralization and collagenization of the nucleus pulposus and extrusion of this material through a ruptured annulus fibrosus resulting in spinal cord compression.1 The MRI appearance of extruded mineralized disks manifests as low signal intensity in spin echo T1-W and T2-W sequences.2
Other causes of an increase in signal intensity in T1-W images include fat, hemorrhage, proteinaceous fluid, melanin, and paramagnetic substances such as iron or manganese.3 In this dog, STIR sequences of the most cranial part of the lesion revealed no loss in signal intensity, ruling out the likelihood of fat while implicating bone or ferromagnetic substances (eg, blood products). Findings on the T2* GE sequences ruled out hemorrhage, and the lack of contrast enhancement made infection or neoplasia unlikely. This combination of findings suggested that hyperintensity in T1-W sequences of the cranial portion was indicative of mineralization or bone and, in this context, was most likely extruded mineralized disk material. The extradural hemorrhage caudal to this hyperintense area was recognized as mixed signal intensity in T1-W and T2-W sequences, confirmed by T2* GE sequences as areas of signal void.
The dog underwent a left-sided hemilaminectomy from L1 to L2. A large amount of compressive extradural material suggestive of mineralized disk, hemorrhage, and epidural fat was obtained. The most cranial compartment was mineralized disk material, the middle compartment consisted of epidural fat, and the most caudal compartment contained hemorrhagic material. Histologic examination of the material from the most cranial compartment revealed mineralized fibrocartilaginous material intertwined with hemorrhagic material and strands of fibrous connective tissue consistent with extruded mineralized nucleus pulposus and ruptured annulus fibrosus. The dog made an uneventful recovery and was clinically normal 2 months later.
- 1.↑
Macias C, Mckee WM, May C, et al. Thoracolumbar disc disease in large dogs: a study of 99 cases. J Small Anim Pract 2002;43:439–446.
- 2.↑
Besalti O, Pekcan Z, Sirin YS, et al. Magnetic resonance imaging in dogs with thoracolumbar intervertebral disk disease: 69 cases (1997–2005). J Am Vet Med Assoc 2006;228:902–908.
- 3.↑
Cakirer S, Karaarslan E, Arslan A. Spontaneously T1-hyperintense lesions of the brain on MRI: a pictorial review. Curr Probl Diagn Radiol 2003;32:194–217.
