Introduction
A 9-year-old 38.7-kg spayed female Labrador Retriever was referred for evaluation of pelvic limb weakness that had progressed to paraplegia over approximately 10 days. The dog’s initial clinical signs included ataxia, knuckling, and crossing of the pelvic limbs. The primary veterinarian treated the dog with prednisone (0.5 mg/kg, PO, q 12 h) and a single injection of polysulfated glycosaminoglycan (Adequan). The dog’s clinical signs improved initially but quickly deteriorated 3 days after this treatment. The patient was mostly an outdoor dog.
Assessment
Anatomic diagnosis
The paraplegia was most likely a result of T3-L3 myelopathy, and the absent proprioceptive positioning of the pelvic limbs could have been due to a lesion involving the forebrain, thalamus, or spinal cord. Rule-out locations for the absent superficial pain sensation in the right pelvic limb and tail and diminished superficial pain sensation in the left pelvic limb included sensory receptors of the skin (Aδ fibers), the spinal cord (lateral cervical nucleus, spinothalamic tract, and spinocervicothalamic tract), the brainstem, the thalamus, and the cerebral cortex (somatosensory cortex). The midthoracic cutoff of the cutaneous trunci reflex was most likely attributable to a T3-L3 myelopathy.
Likely location of a single lesion
A T3-L3 myelopathy was the most likely cause of the dog’s clinical signs.
Etiologic diagnosis
Differential diagnoses considered for a progressive, painful thoracolumbar myelopathy with an acute onset in an outdoor, senior dog included inflammatory or infectious disease, intervertebral disk disease, and neoplasia.
Diagnostic Plan
The diagnostic plan included a CBC and serum biochemical profile to evaluate systemic health, thoracic radiography, advanced imaging (CT or MRI) of the spine, CSF analysis, and infectious disease serologic testing, if indicated.
Diagnostic Test Findings
Results of a CBC and serum biochemical profile were unremarkable except for a moderately high alkaline phosphatase activity (792 U/L; reference range, 23 to 212 U/L). Thoracic radiography revealed chronic thickening of the lower airways, a diffuse bronchial pattern, and pleural fibrosis. No distinct soft tissue pulmonary nodules, intrathoracic lymphadenopathy, or pleural effusion were identified. Severe mid and caudal thoracic spondylosis deformans was present.
Magnetic resonance imaging was performed with a commercial unit (Signa Excite 1.5T; GE Healthcare); sequences obtained consisted of multiplanar T2-weighted, FLAIR, T2*-weighted gradient echo, and pre- and postcontrast T1-weighted images. Tissue in the L1-2 intervertebral disk space and along the ventral and left ventral aspects of the vertebral canals of L1 and L2 was hyperintense on T2-weighted images with peripheral contrast enhancement (Figure 1). There was severe left ventrolateral spinal cord compression over the body of L1, moderate ventral spinal compression over the body of L2, and more focal right ventral spinal cord compression at the level of the L2-3 intervertebral disk. The T9 and T10 vertebral endplates, L1 and L2 vertebral bodies, L4 vertebral body, cranial L5 endplate, and L6 vertebra were hyperintense on T2-weighted images and contrast enhancing, and the T10-11, T11-12, T12-13, T13-L1, and L6-7 intervertebral disks were hyperintense on T2-weighted images. A hyperintense intramedullary spinal cord signal was noted at T10 and T11 and T13-L1 on T2-weighted images. Contrast enhancement of the paraspinal musculature of the caudal thoracic and cranial lumbar spinal regions was noted. Contrast enhancement of several large periaortic and medial iliac lymph nodes was also noted.
Midsagittal T2-weighted (A), transverse T2-weighted (B), midsagittal postcontrast T1-weighted (C), and transverse postcontrast T1-weighted (D) MRI images of the thoracolumbar spine of a 9-year-old Labrador Retriever evaluated because of pelvic limb weakness that had progressed to paraplegia over approximately 10 days. Notice the hyperintense tissue within the L1-2 intervertebral disk space on the midsagittal T2-weighted image (A) with peripheral contrast enhancement and hyperintensity of the L1 and L2 vertebral bodies on T1-weighted images (circle; C). The hyperintense tissue protruded into the neural canal, resulting in severe compression of the spinal cord (arrows; A, B, and D). There was also contrast enhancement of the paraspinal musculature (arrowheads; D).
Citation: Journal of the American Veterinary Medical Association 260, 11; 10.2460/javma.21.04.0224
The MRI findings were consistent with multifocal, severe diskospondylitis with osteomyelitis, severe meningitis, myelitis, myositis, cellulitis, and compressive empyema at the level of L1 and L2. Infected soft tissues and disk herniation causing spinal cord compression at T13-L1 and L2-3 were also noted. Sites of diskospondylitis and osteomyelitis included T10-11, T11-12, T12-13, T13-L1, and L6-7 with osteomyelitis at T10, T11, L1, L2, L4, and L5.
Treatment
A left-sided hemilaminectomy at T13-L1, L1-2, and L2-3 was performed immediately following MRI. A large amount of mucopurulent discharge drained from the vertebral canal at T13-L1, with some additional mucopurulent material noticed at the level of L2-3. The discharge did not appear to be present within the dura during surgery. Samples collected directly from the surgical site were submitted for bacterial culture and susceptibility testing. The entire area was copiously lavaged with sterile saline (0.9% NaCl) solution and evaluated for hemorrhage. The spinal cord appeared decompressed at the end of surgery. A hemostatic sponge (Gelfoam) soaked in sterile saline solution was placed over the hemilaminectomy site prior to closure.
Immediate postoperative treatment consisted of enrofloxacin (10 mg/kg, IV, q 24 h), ampicillin-sulbactam (22 mg/kg, IV, q 8 h), gabapentin (8 mg/kg, PO, q 8 h), prednisone (0.5 mg/kg, PO, q 12 h), methocarbamol (20 mg/kg, PO, q 8 h), a fentanyl continuous rate infusion (3 μg/kg/h, IV), and IV fluid therapy at a maintenance rate. A fentanyl transdermal patch (100 μg) was also applied. The dog was weaned off the fentanyl continuous rate infusion the following day. Motor function in the pelvic limbs was noticed 2 days after surgery. Therefore, basic rehabilitation exercises, including passive range of motion and standing exercises, were added to the treatment regimen to help maintain mobility and strengthen the muscles. Antimicrobials were transitioned from IV to PO administration 5 days after surgery. The dog was discharged on the sixth day after surgery with prescriptions for amoxicillin-clavulanate (23 mg/kg, PO, q 8 h), enrofloxacin (5 mg/kg, PO, q 24 h), prednisone (0.5 mg/kg, PO, q 12 h for 10 days, then 0.5 mg/kg, PO, q 24 h), gabapentin (8 mg/kg, PO, q 8 h), and omeprazole (1 mg/kg, PO, q 24 h). Results of serologic testing for Blastomyces dermatitidis, Histoplasma capsulatum, Coccidioides spp, and Aspergillus spp were negative. Staphylococcus pseudintermedius was isolated from samples collected from the surgical site, and cefpodoxime was added to the treatment regimen on the basis of results of susceptibility testing.
During the 2-week postoperative recheck examination, the dog had ambulatory paraparesis with mild to moderate ataxia. The dosage of prednisone was tapered beginning the fifth week after surgery (0.5 mg/kg, PO, q 48 h for 7 days, then discontinued). Results of physical and neurologic examinations 5 months after surgery were normal. Nine months after surgery, the patient was reportedly doing well and had returned to normal activities at home. At that time, the dog was receiving amoxicillin-clavulanate (20 mg/kg, PO, q 8 h), cefpodoxime (5 mg/kg, PO, q 24 h), and gabapentin (7 to 14 mg/kg, PO, q 8 h, as needed).
Comments
Spinal epidural empyema is an uncommon yet serious disease resulting from a septic and suppurative process in the epidural space of the vertebral canal. It usually develops secondary to a distal or local infection through hematogenous circulation, contiguous spread, direct inoculation, foreign body migration, or parasitic infection. The infectious organisms can originate from the urogenital tract, heart valves (eg, endocarditis), or skin.1,2
Large- or giant-breed dogs appear to have a higher risk of developing spinal epidural empyema, although the condition has been reported in multiple breeds and species.3 Pregnancy could be a risk factor because prolonged exposure to progesterone suppresses the immune system.4 Younger age may also be a risk factor given that 4 cats in a previous report1 were all < 3 years old. Younger animals are frequently more active, making them more susceptible to fight wounds and traumatic injury.
Commonly reported clinical signs of spinal epidural empyema include anorexia, lethargy, fever, hyperesthesia, and progressive neurologic deficits.5 A CBC often reveals leukocytosis characterized by neutrophilia,1–8 and the CSF may have a moderate to high protein concentration with lymphocytic and neutrophilic pleocytosis.1,6,9,10
Magnetic resonance imaging is the diagnostic modality of choice for dogs suspected to have spinal epidural empyema because it provides greater detail of nervous tissue than CT. The lesion in dogs with spinal epidural empyema usually appears as a well-defined epidural mass within the vertebral canal causing variable degrees of spinal cord compression. It is hyperintense or of mixed signal intensity on T2-weighted images, heterogeneously hyperintense on FLAIR images, and hypointense to mildly hyperintense on T1-weighted images. Focal areas of signal void resulting from hemorrhage may be noticed on T2*-weighted images. Rim enhancement is commonly seen on postcontrast images, with diffuse enhancement being present less often. An increased signal intensity in the gray matter of the adjacent spinal cord can be seen. Concurrent soft tissue inflammation (eg, myositis, tendonitis, and meningitis), diskospondylitis, and osteomyelitis are often observed.3
The lesions of spinal epidural empyema can be difficult to discern on CT images. Computed tomography findings suggestive of spinal epidural empyema include a mass effect in the epidural space accompanied by pathologic changes of the bones. Computed tomography myelography may facilitate lesion localization; however, there may be a risk of iatrogenic infection because of percutaneous injection of contrast material.
Staphylococcus and Streptococcus spp are the most common causative bacteria in dogs with spinal epidural empyema. Other possible causative organisms include Bacteroides spp, Clostridium perfringens, Corynebacterium spp, Enterococcus faecalis, Escherichia coli, Klebsiella pneumonia, Pasteurella spp, Corynebacterium spp, Prevotella spp, and Sphingomonas paucimobilis.1
Given the accumulation of purulent material in the epidural space, spinal epidural empyema is deemed a neurologic emergency that requires immediate surgical decompression.10 However, satisfactory outcomes and rapid recoveries have been reported in some patients, including some with severe neurologic deficits, treated without surgery.5 Treatment consists of administering broad-spectrum antimicrobials and potent analgesics. Cerebrospinal fluid analysis is generally not recommended given the low yield and theoretical risk of spreading infection from the epidural space into the subarachnoid space.1,4
References
- 1. ↑
Guo S, Lu D. Clinical presentation, diagnosis, treatment and outcome of spinal epidural empyema in four cats (2010 to 2016). J Small Anim Pract. 2020;61(6):381–388. doi:10.1111/jsap.12943
- 2. ↑
Woodruff M, Rosenblatt AJ, Punke J, Heading K. Concurrent spinal epidural empyema and endocarditis in a dog. Can Vet J. 2019;60(11):1171–1176.
- 3. ↑
De Stefani A, Garosi LS, McConnell FJ, Diaz FJ, Dennis R, Platt SR. Magnetic resonance imaging features of spinal epidural empyema in five dogs. Vet Radiol Ultrasound. 2008;49(2):135–140. doi:10.1111/j.1740-8261.2008.00339.x
- 4. ↑
Gemmill TJ. What Is Your Diagnosis? Epidural empyema. J Small Anim Pract. 2008;49(2):110–112. doi:10.1111/j.1748-5827.2007.00498.x
- 5. ↑
Monteiro SR, Gallucci A, Rousset N, et al. Medical management of spinal epidural empyema in five dogs. J Am Vet Med Assoc. 2016;249(10):1180–1186. doi:10.2460/javma.249.10.1180
- 6. ↑
Levshin S, Davies ES, Van Hatten R, Williamson BG. What Is Your Neurologic Diagnosis? J Am Vet Med Assoc. 2017;251(7):787–790. doi:10.2460/javma.251.7.787
- 7.
Rapoport K, Shamir MH, Bibring U, Barnoon I, Shipov A, Chai O. Epidural spinal empyema and vertebral osteomyelitis in a cat. Isr J Vet Med. 2016;71(4):41–44.
- 8. ↑
Maeta N, Kanda T, Sasaki T, Morita T, Furukawa T. Spinal epidural empyema in a cat. J Feline Med Surg. 2010;12(6):494–497. doi:10.1016/j.jfms.2010.01.015
- 9. ↑
Granger N, Hidalgo A, Leperlier D, et al. Successful treatment of cervical spinal epidural empyema secondary to grass awn migration in a cat. J Feline Med Surg. 2007;9(4):340–345. doi:10.1016/j.jfms.2007.01.004
- 10. ↑
Lavely JA, Vernau KM, Vernau W, Herrgesell EJ, LeCouteur RA. Spinal epidural empyema in seven dogs. Vet Surg. 2006;35(2):176–185. doi:10.1111/j.1532-950X.2006.00129.x
