Fungal infections of the spine are uncommon and usually result from
hematogeneous spread of a primary
focus1.
Blastomyces dermatitidis has been described as a causative agent in
cases of vertebral
osteomyelitis2,3.
Most cases of blastomycosis osteomyelitis begin as primary fungal infections
of the lungs3.
However, traumatic innoculation, local extension, and lymphatic spread also
have been described as possible mechanisms of introduction of the fungi to
bone2. The long
bones are more commonly affected than the vertebral column
is4. Primary
cutaneous blastomycosis can
occur5, and many
other organ systems, including the genitourinary and central nervous systems,
can be
affected6.
Blastomyces dermatitidis is a dimorphic fungus endemic to the
southeastern and south central portions of the United
States4. Primary
pulmonary blastomycosis is acquired by inhalation of airborne
conidia1. The spores
then develop into the yeast form, producing an acute illness stage that is
clinically indistinguishable from a bacterial pneumonia. In as many as 50% of
cases of disseminated blastomycosis, osseous involvement may
occur2,7.
We report the case of a patient in whom vertebral blastomycosis
osteomyelitis was successfully treated nonoperatively. The patient was
informed that information concerning this case would be submitted for
publication.
Asixty-four-year-old South American woman who had recent-onset midthoracic
spinal pain with left scapular radiation presented to us in February 2001. The
patient had no history of trauma or recent illness and no history of exposure
to farm animals or pets. The medical history included chronic hypertension,
arthritis, and depression. Physical examination revealed normal findings.
A computerized tomography scan demonstrated a paraspinal mass with rim
enhancement from T3 through T10 as well as partial focal destruction of the
vertebral bodies on the left side. The patient was then lost to follow-up for
several months.
On July 4, 2001, the patient was readmitted to the hospital. Physical
examination, including a detailed neurologic examination, revealed normal
findings. The midthoracic region of the spine was tender to palpation. The
erythrocyte sedimentation rate was 97 mm/hr (normal, 0 to 20 mm/hr).
Tuberculin skin-testing was negative, and no antibiotic treatment was
initiated pending the results of a biopsy of the paraspinal mass. A
computerized tomography-guided biopsy of the paraspinal mass at T5-T6 was
performed on July 9, at which time three large core-biopsy specimens and two
fine-needle aspiration specimens were obtained and sent for pathological
analysis, culture, sensitivity testing, and polymerase chain reaction for
tuberculosis. The polymerase chain reaction was negative for tuberculosis, and
the findings of the histopathological studies and cultures were
nondiagnostic.
On July 20, a repeat biopsy of the paraspinal mass was performed. Treatment
with oral azithromycin was empirically started, and the patient was discharged
from the hospital while awaiting the results of culture.
On July 31, the findings of physical examination remained normal and the
erythrocyte sedimentation rate was 137 mm/hr. Magnetic resonance imaging
revealed a paraspinal abscess and mild-to-moderate spinal cord compression
(Figs. 1-A and 1-B). Cultures
of specimens obtained during the second computerized tomography-guided biopsy
were positive for an unidentified species of yeast. The histopathological
findings were again nonspecific, and there were no granulomas. Treatment with
azithromycin was stopped, and the patient was started on Abelcet for the
treatment of suspected blastomycosis osteomyelitis and the paraspinal
abscess.
The Abelcet treatment was continued for four weeks, at which time the
erythrocyte sedimentation rate had decreased to 37 mm/hr. Microbiological
studies eventually revealed that the yeast in question was Blastomyces
dermatitidis. On August 29, a repeat computerized tomography scan
revealed that the abscess had decreased in size. Neurologic function remained
intact throughout the course of hospitalization. After four weeks of Abelcet
treatment, treatment with oral itraconazole (400 mg per day) was started and
was continued for a total of six months.
Fourteen months after discharge and thirteen months after the diagnosis,
the patient returned for routine follow-up. She had no complaints and no spine
pain. The erythrocyte sedimentation rate was 24 mm/hr, and physical
examination revealed normal findings. A repeat magnetic resonance imaging scan
showed complete resolution of the anterior paraspinal abscess and soft-tissue
mass and complete resolution of the spinal cord impingement that had been
noted on the initial pretreatment magnetic resonance images
(Fig. 2).
Fungal infections of the spine are uncommon but not rare, and delays in
diagnosis are common. Frazier et al. noted an average delay in diagnosis of
almost three months in a retrospective review of fungal infections of the
spine that were treated at several
institutions8. In
the case described in the present report, the diagnosis was made six months
after the initial symptoms had begun.
Coccidioides immitis osteomyelitis can present a similar clinical
picture to that seen in association with vertebral
blastomycosis9, but
blastomycosis is more common in the southeast United States and
coccidioidomycosis is more common in the southwest. Various species of
Candida10,11,
Aspergillus, Pseudallescheria, and Cryptococcus have also been described as
causative agents in cases of fungal vertebral
osteomyelitis2,12,13.
The differential diagnosis of granulomatous osteomyelitis is broad. It
includes noninfective diseases such as sarcoidosis and metastatic cancer and
various infectious
etiologies2. Fungal
infections should always be ruled out. However, Mycobacterium species,
Brucella, Actinomyces, and Nocardia have also been described as causing
lesions that are histologically and radiographically similar to fungal
osteomyelitis (that is, lytic lesions with or without sclerotic
borders)2-9,14.
Thus, biopsy specimens should be sent for both histopathological analysis and
cultures for the aforementioned bacteria, fungi, and mycobacteria.
In the case described here, the patient was managed nonoperatively with
Abelcet followed by oral itraconazole. Hadjipavlou et al. also reported the
case of a patient in whom lumbar spine blastomycosis was successfully treated
with amphotericin B followed by oral
itraconazole7.
Saccente et al. reported on eight patients with vertebral blastomycosis and
concluded that surgical débridement was not usually
needed3.
Frazier et al. favored surgical treatment in their long-term follow-up
study of eleven patients with fungal discitis and vertebral
osteomyelitis8. The
authors found that all of the patients required surgical treatment despite the
fact that ten of the patients had been managed with amphotericin B. In that
series, however, none of the patients had blastomycosis. Eight of the eleven
patients also had development of paralysis which, by itself, normally would
require surgical treatment.
Amphotericin B and Abelcet are both effective antifungal agents but are
more toxic systemically than the azoles are. Conaughty et al., in a rabbit
model, found that that fluconazole, a member of the azole family (cytochrome
P-450 enzyme inhibitors), had superior penetration into the nucleus pulposus
as compared with amphotericin B and
Abelcet15. These
data imply that itraconazole, a molecule of similar size and polarity, most
likely also has superior penetration into the nucleus pulposus. Currently,
itraconazole is the recommended first-line azole and may be used in lieu of or
after amphotericin
B3.
On the basis of current information, we recommend that vertebral
blastomycosis should initially be treated nonoperatively in the absence of
neurologic deficits and/or a major kyphotic deformity. Large paraspinal fluid
collections may be drained percutaneously. If neurologic deficits, major
kyphotic deformity, or spinal instability are present, surgical treatment
should be performed. ?