Veterinary Surgery

37:801–808, 2008

Spinal Epidural Abscess in Two Calves

DAVIDE D. ZANI, DVM, PhD, LAURA ROMANÒ, DVM, MASSIMILIANO SCANDELLA, DVM, PhD,
MARCO RONDENA, DVM, PhD, PIETRO RICCABONI, DVM, PhD, NICOLA MORANDI, DVM,
ROCCO LOMBARDO, DVM, PhD, Diplomate ACVIM (Neurology), Diplomate ECVN, MAURO DI GIANCAMILLO, DVM,
ANGELO G. BELLOLI, DVM, and DAVIDE PRAVETTONI, DVM, PhD

Objective—To report clinical signs, diagnostic and surgical or necropsy findings, and outcome in 2
calves with spinal epidural abscess (SEA).
Study Design—Clinical report.
Animals—Calves (n ¼ 2).
Methods—Calves had neurologic examination, analysis and antimicrobial culture of cerebrospinal
fluid (CSF), vertebral column radiographs, myelography, and in 1 calf, magnetic resonance imaging
(MRI). A definitive diagnosis of SEA was confirmed by necropsy in 1 calf and during surgery and
histologic examination of vertebral canal tissue in 1 calf.
Results—Clinical signs were difficulty in rising, ataxia, fever, apparent spinal pain, hypoesthesia, and
paresis/plegia which appeared 15 days before admission. Calf 1 had pelvic limb weakness and
difficulty standing and calf 2 had severe ataxia involving both thoracic and pelvic limbs. Extradural
spinal cord compression was identified by myelography. SEA suspected in calf 1 with discos-
pondylitis was confirmed at necropsy whereas calf 2 had MRI identification of the lesion and was
successfully decompressed by laminectomy and SEA excision. Both calves had peripheral neutroph-
ilia and calf 2 had neutrophilic pleocytosis in CSF. Bacteria were not isolated from CSF, from the
surgical site or during necropsy. Calf 2 improved neurologically and had a good long-term outcome.
Conclusion—Good outcome in a calf with SEA was obtained after adequate surgical decompression
and antibiotic administration.
Clinical Relevance—SEA should be included in the list of possible causes of fever, apparent spinal
pain, and signs of myelopathy in calves.
r Copyright 2008 by The American College of Veterinary Surgeons

INTRODUCTION Treatment usually involves surgical decompression with

or without drainage of the abscess, followed by prolonged

I NFECTIONS INVOLVING the epidural space are

rare, however, they represent a devastating neurosur-
gical emergency.1–3 In humans, spinal epidural abscess
(SEA) or spinal epidural empyema (SEE) occurs in 0.2–
1.3 per 10,000 hospital admissions and typically affects
immunosuppressed patients.1,3,4 Clinical presentation and
course of SEA vary from subtle to dramatic. Clinical
findings depend on the position and extension of the ab-
scess or empyema. Despite a significant improvement in
outcomes, there is still a 15% mortality rate, and 38% of
human patients have persistent neurologic dysfunction.5
antimicrobial therapy. If the abscess extends over many
segments, or extends in a panspinal fashion, surgical
treatment can involve multilevel destabilizing laminec-
tomy that requires stabilization of the spinal column.3,6
SEA and SEE have been described in dogs, which like
humans have a poor prognosis despite appropriate ther-
apy. This is often because the delay in diagnosis can result
in progression of clinical signs.7–10 In dogs, SEE is sus-
pected by neurologic examination and spinal diagnostic
imaging, but definitive diagnosis is made during surgery
and by histopathology or necropsy.7 The most common

From the Dipartimento di Scienze Cliniche Veterinarie and Dipartimento di Patologia Animale Igiene e Sanità Pubblica Veterinaria,
Ospedale Veterinario Grandi Animali, University of Milan, Lodi, Italy.
Address reprint requests to Dr. Davide Pravettoni, DVM, PhD, Dipartimento di Scienze Cliniche Veterinarie, Clinica dei Ruminanti e
del Suino, University of Milan, Via dell’Università, 6 Lodi LO 26900, Italy. E-mail: davide.pravettoni@unimi.it.
Submitted April 2007; Accepted December 2007
r Copyright 2008 by The American College of Veterinary Surgeons

0161-3499/08
doi:10.1111/j.1532-950X.2008.00454.x
801
802 SPINAL EPIDURAL ABSCESS
clinical signs are lethargy, fever, anorexia, spinal pain,
and paraparesis or paraplegia.10,11 Common laboratory
abnormalities are peripheral neutrophilia and neutro-
philic pleocytosis in cerebrospinal fluid (CSF).7,8 SEA is
not always directly visible on survey radiographs. Lateral
and sagittal views can show associated abnormalities
such as concurrent discospondylitis, vertebral bone pro-
liferation, and vertebral luxation.7,9 Myelography, com-
puted tomography and magnetic resonance imaging
(MRI) are necessary to recognize focal, multifocal, or
diffuse spinal cord compression7,9,11 and identify the site
and the nature of the lesion.12–14
Vertebral infections involving the spinal cord which
cause neurologic disorders are also reported in rumi-
nants. Udall15 described a cow with an abscess involving
T8–T9 that led to a dog-sitting posture and the animal’s
inability to stand. Robertson and Boucher16 reported an
abscess involving T13 and L1 that caused pelvic limb
paresis. Sherman and Ames17 described 5 cattle with ver-
tebral body abscesses. More recently Braun et al18 de-
scribed clinical and laboratory findings in 11 cattle with
an abscess involving the thoracic vertebrae and 4 cattle
with an abscess in the cervical vertebrae.19 These condi-
tions led to progressive paresis and paralysis of the pelvic
limbs or tetraparesis when the lesion was cranial to
T2.18,19 The most important hematologic findings were
an increased concentration of total plasma proteins and
fibrinogen or reduced clotting time in the glutaraldehyde
test. Analysis of the CSF showed slightly increased pro-
tein concentration.
Vertebral body abscess that extended into the epidural
space causing spinal cord compression was diagnosed in
the cervical and thoracic spinal cord in 8 lambs.20 These
lambs were 4–10 weeks old, with locomotor dysfunction
or paresis and a significant increase in protein concen-
tration in lumbar CSF samples compared with cisternal
CSF samples. Two lambs also had neutrophilic pleocyto-
sis in lumbar CSF samples. Epidural abscess involving
the thoracolumbar spinal cord was diagnosed in 5 sheep
with pelvic limb paresis with an increase in the protein
concentration and in total white blood cell and neutro-
phil percentage in lumbar CSF.21
We report clinical findings and diagnostic procedures
used in 2 calves with SEA. Definitive diagnosis was per-
formed after necropsy in calf 1 and during decompressive
laminectomy and histologic examination of tissue from
the vertebral canal in calf 2.
CLINICAL REPORT
Calf 1
A 3.5-month-old male Holstein Friesian–Limousine
cross calf was admitted for evaluation of pelvic limb
weakness, paresis, and ataxia that had progressively
worsened over 15 days and was associated with poor
weight gain. Pneumonia diagnosed 1 month earlier had
been treated by intramuscular administration of procaine
benzylpenicillin (20,000 U/kg) and dihydrostreptomycin
sulfate (25 mg/kg). On admission, the calf was in fair
physical condition, with normal-sized lymph nodes, in-
creased heart (120 beats/min) and respiratory
(108 breaths/min), body temperature of 39.41C, regular
ruminal motility, and normal fecal appearance.
The calf was alert, had an abnormal posture when
lying down, characterized by frog leg positioning of the
hind limbs, and would attempt to stand but was unable to
do so unassisted. When standing, the calf was reluctant to
walk and had severe ataxia. No forelimb neurologic defi-
cits were detected but hind limb proprioception was re-
duced. Patellar, cranial tibial, and gastrocnemius tendon
reflexes, tested in lateral recumbency, were normal. Sen-
sitivity to a cutaneous sensory test was gradually reduced
starting from the thoracic region to the hind limbs.
Total white blood cell count was increased (13.26 109/L
[reference interval: 4–10 109/L]; 34.9% neutrophils, 54.8%
lymphocytes, 8.1% monocytes) as were creatine kinase
(CK; 191 U/L; reference value o100 U/L) and lactate
dehydrogenase (LDH; 2424 U/L; reference value
o1500 U/L). Lumbar CSF was normal and no organisms
were isolated on microbial culture.
On lateral survey radiographs of the thorax there was
a mixed interstitial/alveolar pulmonary pattern with air
bronchograms within the cranial lobes consistent with
chronic bronchopneumonia. A circular filling defect was
observed at the apophysis of T4. An oval shaped, well
defined, 9 cm
10 cm region with soft tissue radiopacity
was identified beneath T9–T11 vertebral bodies, associ-
ated with collapse of the T9–T10 intervertebral space, end
plate destruction with shortening of the vertebral bodies,
and subluxation of T9. These radiographic changes were
consistent with discospondylitis.
Cervical myelography was performed with the calf
anesthetized and positioned in right lateral recumbency.
s
After contrast medium injection (Iohexol–Omnipaque
300, Amersham Health S.r.l., Milan, Italy; 0.3 mL/kg at
3 mL/min), the table was raised 251 to elevate the head
and neck and facilitate caudal progression of contrast.
Lateral projections of the spine were taken at 0, 7, 15, and
20 minutes after administration. Failure in contrast pro-
gression was observed at T9–T10 (Fig 1A). To further
evaluate the compressive lesion, the calf was repositioned
in a lateral recumbency and contrast was injected in the
lumbosacral space. Severe attenuation of contrast was
observed at T10 with dorsal displacement of the spinal
cord at T9–T10 (Fig 1B).
Clinical findings, survey radiographs, and myelography
were consistent with a diagnosis of discospondylitis
 ZANI ET AL
Fig 1. Calf 1—thoracic myelogram, lateral projection. Note
the failed progression of contrast at T9 (A) and severe atten-
uation of contrast at T10 during a combined myelogram (B).
The dorsal displacement of the spinal cord is caused by sub-
luxation of T9 (a) associated with collapse of T9–T10 inter-
vertebral space (a1) and shortening of the vertebral bodies.
Under T9–T11 vertebral bodies there is a soft tissue radi-
opaque oval-shaped area of diameter 9 cm
10 cm (b) and a
circular defect can be observed at the apophysis of T4 (c).
with severe extradural spinal cord compression at T9–T10.
The calf was euthanatized at the owner’s request. At nec-
ropsy there was an abscess localized in the T4 vertebral
apophysis. Multiple, linear, encapsulated chronic abscesses
(1–2.5 cm diameter) with a necrotic liquefactive center,
originated from this lesion, and extended through the left
epaxial dorsal musculature to T9–T10 where they invaded
the hypaxial musculature. An encapsulated abscess (8 cm
diameter) ventral to the T9–T10 vertebral bodies extended
into the thoracic cavity at the caudal edge of the right
caudal lung lobe. The abscess involved the intervertebral
disk causing discospondylitis, invaded the spinal canal,
and compressed the dura mater and spinal cord. The dura
mater appeared thickened and affected by mild fibrosis;
the arachnoid spaces were dilated cranial to the lesion and
intense meningeal edema and vascular congestion were
present. Histologically, the foci were encapsulated by a
thick fibrovascular layer, surrounded by epithelioid mac-
rophages, foamy cytoplasm macrophages, lymphocytes,
plasma cells, neutrophils and small numbers of giant mul-
tinucleate foreign-body-type cells and eosinophils. The ex-
tensive center of coagulative necrosis was characterized by
necrotic and cellular debris mixed with karyolytic neutro-
phils. No bacteria were recognized in hematoxylin and
eosin and modified Gram stain sections, suggesting a
histologic diagnosis of sterile chronic pyogranulomatous
inflammation. The compressed tract of the spinal
cord and the neighboring segment were characterized
by severe axonal degeneration, initial demyelination of
white fibers, neuronal necrosis, and gliosis suggesting
Wallerian degeneration. Neither meningitis nor myelitis
were identified in the histologic sections. The lesions
observed were the most likely cause of the neurologic signs
803
and were similar to those previously described for
SEA.2,7,10,22
Calf 2
A 2-month-old male Holstein Friesian–Belgian Blue
cross calf was admitted with a 15-day history of progres-
sive difficulty in standing-up that began when the calf was
recovering from moderate enzootic pneumonia, treated
with marbofloxacin (2 mg/kg intramuscularly [IM]). The
calf had recently been immunized against Clostridium
perfringens infection.
On clinical examination, the calf was in good physical
condition with a good appetite. Body temperature was
39.61C, breathing was costoabdominal (36 breaths/min),
pulse was rhythmic (104 beats/min), and ruminal motility
and defecation were normal. The calf could only stand
when assisted and was reluctant to move, with severe
ataxia involving both fore and hind limbs. When stand-
ing, the calf rapidly lost balance and collapsed while
walking. On neurologic examination postural reactions
were decreased in all limbs but mental status and cranial
nerve function was normal. There was no evidence of
decreased tone, muscle atrophy or decreased spinal re-
flexes in fore or hind limbs. Passive neck movements were
reduced and elicited a painful reaction. The clinical signs
were consistent with a cervical spinal cord lesion rostral
to the cervicothoracic intumescence.
White blood cell count was 8.14
109/L (reference
interval, 4–10
109/L) with 51.1% neutrophils, 34.1%
lymphocytes, 8.1% monocytes, and 3.7% eosinophils).
Serum CK (226 U/L) and LDH (2008 U/L) were slightly
increased. Lumbar CSF had rare neutrophils with
pleocytosis, monocytes, and normal total protein con-
centration (22.4 mg/dL) but no microorganisms were iso-
lated on culture.
Survey radiographs of the spine taken after sedation
with xylazine (0.02 mg/kg IM) revealed no abnormalities.
On cervical myelography under isoflurane anesthesia,
deviation and thinning of the subarachnoid contrast me-
dium columns suggestive of ‘‘hour-glass’’ extradural
compression at C3–C4 (Fig 2) were observed on lateral
and ventrodorsal projections. Then the calf was posi-
tioned in lateral recumbency in a 0.2 T open MR system
(Vet MRI, Esaote S.p.A.–Genoa, Italy) with the neck
positioned in a wrap-around receiving coil. Sequences
included sagittal and dorsal T1-weighted images before
and after intravenous CM administration (gadopentetic
acid dimegluminic salt 469 mg/mL at 0.2 mL/kg in bolus
s
IV; Magnevist , Schering AG, Berlin, Germany; time to
repetition TR 800 ms, time to echo TE 26 ms, 3.0 mm
slices, 0.3 mm gaps and TR 690 ms, TE 26 ms, 3.0 mm
slices, 0.3 mm gaps, respectively) and transverse and sag-
ittal T2-weighted images (TR 2800 ms, TE 80 ms, 3.5 mm
804 SPINAL EPIDURAL ABSCESS
Fig 2. Calf 2—Cervical myelogram. (A) Lateral projection.
Marked thinning of the dorsal and ventral subarachnoid con-
trast medium columns is observed at C3–C4. (B) Ventrodorsal
projection. Note an ‘‘hour-glass’’ extradural cord compression
at C3–C4.
slices, 0.3 mm gaps and TR 3000 ms, TE 80 ms, 3.0 mm
slices, 0.3 mm gaps, respectively).
The precontrast T1-weighted images in the dorsal
plane, had a well-defined ‘‘hour-glass’’ compression of
spinal cord because of 2 heterogeneous isointense areas
(comparison made with the adjacent cord) that sur-
rounded the spinal cord at C3–C4 dorsally to laterally,
causing severe extradural cord compression. On the T2-
weighted sagittal images there was a well-defined heter-
ogeneous isointense area (2.1 cm
1.2 cm), localized in
the C3–C4 inter arch space, that displaced the spinal cord
ventrally. This lesion expanded dorsally in a poorly de-
fined hyperintense area. On postcontrast imaging, there
was rim enhancement of the previously described lesions
(Fig 3). MRI confirmed the presence of an extradural
mass, associated with spinal cord compression at C3–C4
and SEA was suspected.
Laminectomy was performed under isoflurane anes-
thesia to further define and potentially treat the condi-
tion. Electrocardiogram, oxygen saturation, invasive
blood pressure, and end-tidal CO2 concentration were
monitored. The calf was positioned in sternal recumbency
with the fore limbs pulled forward, head and neck ex-
tended, and neck elevated to avoid jugular vein compres-
sion. Ropes were used to secure the limbs and adhesive
tape applied to stabilize the head and thorax. After clip-
ping and aseptic preparation, a dorsal median incision
was made from the occiput to the spinous process of T1.
Subcutaneous tissue and fat overlying the funicular part
of the nuchal ligament were incised on the midline and
the m. trapezius cervicalis on the right side was separated
from the nuchal ligament, from the occiput to T1. A
cleavage plan was identified and the dorsal neck muscles
on the right side were separated from the nuchal ligament
and exposure enhanced by self-retaining retractors. By
palpation the spinous processes of C2–C5 were identified.
The multifidus cervicis muscle was elevated from the ver-
tebral laminae of C3 and C4 and from the articular pro-
cesses using periosteal elevators and scissors.
Gelpi retractors were used to maintain exposure of the
C3 dorsal lamina, C3–C4 interarcuate area, and C4 dor-
sal lamina. A soft tissue, smooth surfaced, rounded,
grayish mass (  2 cm diameter) was identified at the level
of, and dorsal to, the C3–C4 interarcuate ligament, where
it appeared to enter the spinal canal through the liga-
ment. Blunt dissection was attempted to separate the
mass from the surrounding tissues but unfortunately
rupture and contamination of the surgical site with pu-
rulent exudate occurred. After suction and saline (0.9%
NaCl) solution irrigation, a ligature was placed in the
abscess capsule, at the level of the interarcuate ligament,
and the extravertebral component of the mass removed.
After removal of the C3 and C4 spinous processes by
rongeurs, a high-speed burr was used for laminectomy
from mid C3 to mid C4 extending laterally to the level of
the articular processes, which were fully preserved. The
interarcuate ligament was removed using an 11 blade and
forceps, then the thin inner cortical bone was removed
using fine bone rongeurs and Kerrison rongeurs to expose
the spinal canal. The spinal cord appeared compressed
dorsally by an extension of the abscess. The abscess was
removed relatively easily without evidence of adhesions
to dura mater or substantial hemorrhage. After copious
saline solution irrigation, the wound was closed in layers.
No microorganisms were isolated from the exudate
collected intraoperatively. Histologic examination of the
excised mass, revealed an inflammatory process that
deeply infiltrated the dorsal skeletal muscles and adjacent
connective tissue. It consisted of multifocal coalescing
foci, lined by a thick fibrovascular capsule with variable-
sized centers containing karyolytic neutrophils admixed
with abundant cellular debris, and postnecrotic dystro-
phic mineralization. This center was surrounded by
epithelioid macrophages, foamy cytoplasm macrophag-
es, lymphocytes, plasma cells, neutrophils, eosinophils,
and small numbers of giant multinucleate foreign-body
type cells. No bacteria or fungal elements were observed.
The morphologic diagnosis was multifocal to coalescing
chronic sterile pyogranulomatous myositis and fascitis.
Postoperatively, procaine benzylpenicillin (40,000 U/
kg for 10 days), prednisolone acetate (1 mg/kg IM for
4 days, and for 3 times on alternate days), and vitamin
 ZANI ET AL 805

Fig 3. Calf 2—cervical spine magnetic resonance imaging (MRI) on admission. (A) C2–C5 level T1W sagittal MRI pre (A1) and
post- (A2) gadolinium. Spinal epidural abscess (SEA) is identified on the C3–C4 interarch space (arrows). The image demonstrates
rim enhancement of epidural mass after contrast injection. (B) C2–C5 level T2W sagittal MRI. (C) C3–C4 level T2W transverse
MRI. SEA (arrow) surrounds and compresses the spinal cord. Note the heterogeneous pattern of the SEA at this level. (D) C3–C4
level T1W dorsal MRI pre- (D1) and post- (D2) gadolinium. Bilateral circular spinal extradural masses (arrows), with rim
enhancement after CM injection, cause an ‘‘hour-glass’’ compression of the spinal cord.

B1 (10 mg/kg IV for 3 days and for 3 times on alternate days, on MRI there was resolution of the spinal cord
days) were administered. The calf was assisted into a compression (Fig 4). The calf was discharged at 1 month,
standing position and walked 5 times daily. It was able to and continued to grow normally without any residual
stand and walk without assistance on the 21st day. At 30 neurologic deficit.
806 SPINAL EPIDURAL ABSCESS

Fig 4. Calf 2—cervical spine magnetic resonance imaging (MRI) 30 days after surgery. (A) C2–C5 level T1W sagittal MRI pre-
(A1) and post- (A2) gadolinium. Although moderate residual dural and paraspinal tissue enhancement is present after CM injection,
laminectomy resulted in complete spinal cord decompression. (B) C2–C5 level T2W sagittal MRI shows a heterogeneous hypo-
isointense (comparison to spinal cord) irregular area probably represented by fibrous healing tissue secondary to dorsal laminec-
tomy. (C) C3–C4 level T1W transverse MRI. The left external jugular vein (arrow) appears as a hyperintense circular area because
involved by iatrogenic thrombophlebitis. (D) C3–C4 level T1W dorsal MRI pre- (D1) and post-gadolinium (D2). Moderate residual
dural enhancement after CM injection is still present.

DISCUSSION Back or neck pain, fever and neurologic deficits are the
3 most common symptoms.2,5 The rarity of SEA, com-
SEA is an uncommon but important disease in hu- bined with the relatively nonspecific clinical signs, can
mans2,23 and has been reported in dogs7–11 and sheep.20,21 result in a delayed diagnosis that may have a catastrophic
 ZANI ET AL
outcome.2 Neurologic deterioration can progress rapidly
and human patients rarely regain neurologic function if
they are paralyzed for 424–36 hours before appropriate
treatment is initiated.2 Moreover, the nature of the
epidural inflammation, which can be consistent with an
encapsulated abscess with a liquefied purulent core or
with phlegmonous tissue characterized by vascularized
granulation tissue with microabscesses, has been corre-
lated to chronic symptoms and outcome.22 Clinical pre-
sentation with symptoms observed for o16 days has
been associated with a largely purulent epidural accumu-
lation and prognosis can be improved because of the
possibility of draining the pus.24 In our calves, neurologic
signs worsened over 15 days after onset. In both calves,
on CSF analysis protein concentration was normal with a
moderate neutrophilic pleocytosis, demonstrating that
spinal cord infection or inflammation was unlikely and
neurologic signs, at least in calf 2, were caused by direct
extradural compression of the spinal cord.
Calf 1 was euthanatized because of severe discos-
pondylitis and vertebral subluxation whereas calf 2 was
admitted for therapy because of the possibility of drain-
ing the lesion and removing the mass identified on diag-
nostic imaging. In humans, MRI is the fastest and most
accurate imaging method for evaluation and diagnosis of
SEA, and for treatment planning.5,25 MR imaging allows
definition of the abscess area, identification of lesion ex-
tent, and degree of surrounding tissue involvement.7,9,12
Moreover MRI permits differentiation of epidural in-
flammatory masses as ‘‘phlegmon’’ or ‘‘abscess’’ based on
the pattern of contrast enhancement.22 Because of our
magnet’s limited field of view (14 cm), we preferred to
localize the site of compression using myelography. Al-
though myelography is associated with a risk of spread of
the infectious agents into the meninges,26 it allows rapid
identification of the site of extradural cord compression
especially when a long tract of column needs to be in-
vestigated, which then allows selective examination by
MRI.
MR features of SEAs are usually characterized by low
or intermediate intensity on T1W images and high or
intermediate intensity on T2W images (comparisons
made with the adjacent spinal cord).1,9 The appearance
of the fluid portion of the abscess tends to produce an
increased signal on T2W images; granulation tissue and
edema surrounding the collection of pus are seen as non-
homogeneous areas with a mildly increased signal.1 After
contrast injection, the enhancement pattern varies de-
pending on age and consistency (frank abscess or
phlegmonous granulation tissue) of the SEA.1,9
In our calf, there was not a characteristic hyperintense
signal on the T2W sequences, but we observed a heter-
ogeneous isointense signal on T2W and a heterogeneous
isointense signal on T1W, associated with a ‘‘ring effect.’’
807
After contrast injection, a thin peripheral rim enhance-
ment, which is the least common pattern in humans,1
suggested a collection of fluid pus with a surrounding,
enhancing capsule. Isointense signals of SEA on T1W
and T2W images were also reported by Lang et al1 in a
50-year-old woman with an epidural abscess localized
between C3 and C7. In this calf, MR imaging was effec-
tive in defining the extent of infection, and in determining
treatment and planning of the surgical approach.
The origin of SEA in these calves and the potential
source of infection was not definitively identified. Calf 1
had bronchopneumonia, but we speculate that the source
of infection was likely drug injections near the apophysis
of T4, subsequent abscess formation with pus drainage in
the paraspinal site along muscular–fascial planes, verte-
bral involvement, and SEA formation. Calf 2 also had
bronchopneumonia before admission and had been
treated with antibiotics. This calf had recently been im-
munized against Clostridium perfringens infection and lo-
calization of the abscess in the neck, which started the
extradural spinal cord compression, seemed to corre-
spond to the usual site of intramuscular injections in
calves. In both calves, the histologic characteristics of the
lesion were consistent with a chronic sterile pyogranulo-
matous foreign-body lesion, suggesting that SEA in these
calves was likely iatrogenic in origin, a finding also re-
ported in humans.23 Although vertebral body or spinal
cord infection after vaccination in cattle has not been
documented to our knowledge, intramuscular adminis-
tration of clostridial vaccines and some antibiotics can
cause damage so severe that it can identified in beef mus-
cle a month later.27
The most common causative agents reported for SEE
in dogs and for SEA in humans include Staphylococcus
and Streptococcus species.2,10,28 That we failed to isolate
bacteria from these lesions may have been influenced by
prolonged administration of broad-spectrum antibiotics
before the calves were admitted. Gelfenbeyn et al4 re-
ported that in SEA, it is possible to identify the pathogens
in 80% of human patients, but the percentage decreases
significantly in subacute or chronic forms. Bacterial iden-
tification is essential for success with nonoperative treat-
ment.26 The postoperative medical protocol used in calf 2
included administration of corticosteroids, which may
suppress the immune response but may also be beneficial
in reducing perioperative spinal cord edema.2
Given the findings we report, SEA should be included
in the differential diagnoses for bovine myelopathies.
Suspicion should be high when cattle are affected by
painful myelopathy accompanied by fever. Because mus-
cular dystrophy is often considered in the differential di-
agnosis, serum CK and LDH activities were measured
in both calves. Nevertheless, if patients have been in a
prolonged recumbent position even when the problem
808 SPINAL EPIDURAL ABSCESS
is primarily neurologic, CK and LDH activity may be
high because of pressure necrosis of muscle and should
not be misinterpreted as nutritional myodystrophy or
myopathy.17
Although in humans and dogs, rapid diagnosis is es-
sential for successful therapy against the rapid deteriora-
tion of the neurologic clinical condition or death,2,7,10,11
in cattle the disease seems to progress more slowly, but
the outcome is compromised if SEA has associated ver-
tebral involvement. The economic aspects of this clinical
approach can be justified only in carefully selected cases
or animals with high genetic value. Nevertheless, we
found it interesting to consider these clinical cases from a
comparative pathology perspective because it provides
insight into diagnosis and treatment in other large animal
species.
REFERENCES
1. Lang IM, Hughes DG, Jenkins JP, et al: MR imaging ap-
pearances of cervical epidural abscess. Clin Radiol 50:466–
471, 1995
2. Darouiche RO: Spinal epidural abscess. N Engl J Med
355:2012–2020, 2006
3. Rea GL, McGregor JM, Miller CA, et al: Surgical treatment
of the spontaneous spinal epidural abscess. Surg Neurol
37:274–279, 1992
4. Gelfenbeyn M, Goodkin R, Kliot M: Sterile recurrent spinal
epidural abscess in a patient with Crohn’s disease: a case
report. Surg Neurol 65:178–184, 2006; discussion 184
5. Reihsaus E, Waldbaur H, Seeling W: Spinal epidural abscess:
a meta-analysis of 915 patients. Neurosurg Rev 23:175–
204, 2000; discussion 205
6. Leonard J, Kaufman B: Treatment of a holocord epidural
abscess. Case illustration. J Neurosurg 94:179, 2001
7. Lavely JA, Vernau KM, Vernau W, et al: Spinal epidural
empyema in seven dogs. Vet Surg 35:176–185, 2006
8. Cherrone KL, Eich CS, Bonzynski JJ: Suspected paraspinal
abscess and spinal epidural empyema in a dog. J Am Anim
Hosp Assoc 38:149–151, 2002
9. Naughton JF, Tucker RL, Bagley RS: Radiographic diag-
nosis—paraspinal abscess in a dog. Vet Radiol Ultrasound
46:23–26, 2005
10. Dewey CW, Kortz GD, Bailey CS: Spinal epidural empyema
in two dogs. J Am Anim Hosp Assoc 34:305–308, 1998
11. Schmiedt CW, Thomas WB: Spinal epidural abscess in a ju-
venile dog. Vet Comp Orthop Traumatol 18:186–188, 2005
12. Parkinson JF, Sekhon LH: Spinal epidural abscess: appear-
ance on magnetic resonance imaging as a guide to surgical
management. Report of five cases. Neurosurg Focus
17:E12, 2004
13. Parkinson JF, Sekhon LH: Surgical management of spinal
epidural abscess: selection of approach based on MRI ap-
pearance. J Clin Neurosci 11:130–133, 2004
14. Widmer WR, Thrall DE: Canine and feline intervertebral disc
disease, myelography, and spinal cord disease, in Thrall
D.E. (ed): Textbook of Veterinary Diagnostic Radiology
(ed 4). Philadelphia, PA, Saunders, 2002; pp 110–126
15. Udall DH: Paraplegia in a cow. Cornell Vet 18:370–371, 1928
16. Robertson JM, Boucher WB: Vertebral body abscess in a
Heifer. J Am Vet Med Assoc 143:1211–1213, 1963
17. Sherman DM, Ames TR: Vertebral body abscesses in cattle:
a review of five cases. J Am Vet Med Assoc 188:608–611,
1986
18. Braun U, Schweizer G, Gerspach C, et al: Clinical findings in
11 cattle with abscesses in the thoracic vertebrae. Vet Rec
152:782–784, 2003
19. Braun U, Gerspach C, Salis F, et al: Klinische Befunde bei 4
Rindern mit Abszess in der Halswirbelsaeule. Schweiz Arch
Tierheilkd 145:124–128, 2003
20. Scott PR, Penny CD, Murray LD: A field study of eight
ovine vertebral body abscess cases. NZ Vet J 39:105–107,
1991
21. Scott PR, Will RG: A report of Froin’s syndrome in five
ovine thoracolumbar epidural abscess cases. Br Vet J
147:582–584, 1991
22. Tung GA, Yim JW, Mermel LA, et al: Spinal epidural ab-
scess: correlation between MRI findings and outcome. Ne-
uroradiology 41:904–909, 1999
23. Brookman CA, Rutledge ML: Epidural abscess: case report
and literature review. Reg Anesth Pain Med 25:428–431,
2000
24. Danner RL, Hartman BJ: Update on spinal epidural abscess:
35 cases and review of the literature. Rev Infect Dis 9:265–
274, 1987
25. Mackenzie AR, Laing RB, Smith CC, et al: Spinal epidural
abscess: the importance of early diagnosis and treatment.
J Neurol Neurosurg Psychiatry 65:209–212, 1998
26. Rigamonti D, Liem L, Sampath P, et al: Spinal epidural ab-
scess: contemporary trends in etiology, evaluation, and
management. Surg Neurol 52:189–196, 1999; discussion 197
27. George MH, Heinrich PE, Dexter DR, et al: Injection-site
lesions in carcasses of cattle receiving injections at branding
and at weaning. J Anim Sci 73:3235–3240, 1995
28. Darouiche RO, Hamill RJ, Greenberg SB, et al: Bacterial
spinal epidural abscess. Review of 43 cases and literature
survey. Medicine (Baltimore) 71:369–385, 1992