Nocardia brain abscess: Severe CNS infection that needs aggressive management; case report

Abstract

Nocardia brain abscess is a rare central nervous system (CNS) infection that carries a high mortality rate reaching 34% which is considered the highest amongst brain abscesses caused by microorganisms. All available literature is in the form of retrospective studies and small case series. In this case report the authors present a patient whose course of disease was stormy and required multiple neurosurgical procedures. The clinical outcome, long-term follow up and a review of the literature is discussed.

Full text

CASE REPORT
Nocardia brain abscess: severe CNS infection that needs
aggressive management; case report
Amr Zakaria &Sherif Elwatidy &Essam Elgamal
Received: 10 March 2008 / Accepted: 30 May 2008
#Springer-Verlag 2008
Summary Nocardia brain abscess is a rare central nervous
system (CNS) infection that carries a high mortality rate
reaching 34% which is considered the highest amongst
brain abscesses caused by microorganisms. All available
literature is in the form of retrospective studies and small
case series. In this case report the authors present a patient
whose course of disease was stormy and required multiple
neurosurgical procedures. The clinical outcome, long-term
follow up and a review of the literature is discussed.
Keywords Nocardia .Brain abscess .
Decompressive craniectomy .CNS infection
Clinical details
A 40 year old man with autoimmune hepatitis had been
treated with immuno-suppressant drugs (azathioprine and
prednisolone) for one and half years. Five days prior to
presentation, he complained of fever, worsening headache
and vomiting. Upon arrival to King Khalid University
hospital, the patient was febrile (39.5°C), the Glasgow coma
score (GCS) was 15/15 and he had no focal neurological
deficits. There were no signs of meningeal irritation but
fundoscopy revealed bilateral papilloedema. Laboratory
work up showed leucocytosis (WBC 14,000/dl), sedimen-
tation rate (70 mm/h in the first hour) and markedly elevated
C-reactive protein (100). Computerised tomography (CT)
and MRI brain scans, done upon arrival, showed a
hypodense lesion in the right occipital lobe surrounded with
peri-lesional oedema with a faint ring enhancement after
contrast injection and evidence of ventriculitis in the form of
enhancement of the choroid plexus and walls of the
ventricles as well as presence of sediment in the occipital
horns of the lateral ventricles (Fig. 1). The lesion appeared
hyperintense in diffusion weighted images of the MRI scan
together with evidence of cerebritis in the contra-lateral
occipital lobe, (Fig. 2). Emergency burr hole aspiration of
the abscess was done and 20 cc of pus aspirated and sent
for microbiology. The initial result of gram staining was
unusual and showed gram positive bacilli and fungal
hyphae. It was also positive for Ziehl–Neelsen staining
(acid fast bacilli). The patient was prescribed broad
spectrum intravenous antibiotics (ceftriaxone 1 g twice
daily and metronidazole 500 mg three times daily), anti-
tuberculous drugs (rifampicin 600 mg orally once daily,
isoniazide 300 mg orally once daily, pyrazinamide 1.5 mg
orally once daily and ethambutol 1.2 g orally once daily) as
well as antifungal drugs (fluconazole 400 mg once daily).
The final culture grew Nocardia asteroids sensitive to
trimethoprim/sulfamethoxazole. The previous antimicro-
bials were replaced with intravenous trimethoprim/sulfa-
methoxazole (TMP, 400 mg SMX, 2 g three times daily).
The patient developed leucopenia and the antimicrobials
were changed again to a combination of minocycline
(100 mg twice daily) and meropenem (2 g three times
daily).
After aspiration, the patient’s condition deteriorated
further and he became lethargic, the headache got worse
and he was continuously febrile. Follow-up CT brain scan
with and without contrast was repeated on the third post-
operative day which showed adequate drainage of the
abscess but the brain oedema had increased and there was
clear evidence of ventriculitis.
Acta Neurochir
DOI 10.1007/s00701-008-0026-2
A. Zakaria :S. Elwatidy (*):E. Elgamal
Division of Neurosurgery, King Khalid University Hospital,
Riyadh, Saudi Arabia
e-mail: smfwat@yahoo.com

The patient was transferred to the critical care unit and
was kept under close monitoring. On the fourth post-
operative day the GCS dropped suddenly to 4/15 and both
pupils became dilated and non-reactive to light. The patient
was intubated and ventilated and 250 ml of 20% mannitol
was infused. An urgent CT brain scan showed recollection of
the abscess, severe generalised brain oedema, obliteration of
the basal cisterns and fullness of the foramen magnum, but
no appreciable change in the size of ventricles (Fig. 3).
The patient was taken immediately to the operating room
where a bi-frontal decompressive craniectomy was done.
The details of this procedure have been described in
previous reports [6,7]. Intra-operatively, the brain was
found to be severely swollen. An external ventricular drain
was inserted and the CSF was purulent due to ventriculitis.
The drain was connected to a closed system and an intra-
parenchymal probe was inserted to monitor the intracranial
pressure. Post-operatively, assisted ventilation was contin-
ued. Immediately following surgery the pupils became
equal in size and small but not reactive to light. Serial
specimens of infected CSF were sent to the microbiology
laboratory.
After decompressive craniectomy the intracranial pres-
sure was always low (±10 mm Hg). The probe was
removed on the fourth post-operative day and the EVD
removed after 10 days. Tracheostomy was performed after
2 weeks of assisted ventilation. The patient’s condition
showed gradual improvement, the fever subsided after
12 days of antibiotic treatment, the WBC and ESR reduced
and the GCS improved to 10/15 after 4 weeks (E4, M5,
V1). He was weaned off the ventilator after 5 weeks. Serial
CT scans (done on weekly basis) showed progressive
ventricular dilatation indicating active hydrocephalus
(Fig. 4). Three weeks after decompressive craniotomy a
left ventriculo-peritoneal shunt was inserted together with
replacement of the bone flap at the same time. Despite
2 months of antibiotic treatment there was incomplete
resolution of the abscess and a second aspiration was done
which yielded sterile fluid with no bacterial growth. The
tracheostomy tube was removed in the ninth week.
Clinically, the patient remained bed ridden with no
verbal response, and had generalised spasticity. He opened
his eyes spontaneously, flexed limbs to painful stimuli and
was fed through a nasogastric tube. Minocyclin was given
for 6 months whilst intravenous meropenen was continued
for 1 year. The patient was transferred back to the referring
hospital for continuation of antibiotics and a physiotherapy
programme.
At the 1 year follow up visit, the patient came to the
clinic in a wheelchair, could stand and walk a few steps
with assistance and uttered simple words. Higher mental
function, such as memory, attention, language skills,
reasoning, problem solving and emotional state were
improving. He could eat and drink independently but
required assistance in other daily activities. He had retired
from his previous job as a teacher. The follow up CT scan
after 1 year demonstrated a healthy bone flap and a
functioning CSF shunt (Fig. 5).
Fig. 2 MRI, diffusion weighted image, showing hyperintensity in
both occipital lobes indicating diffuse cerebritis
Fig. 1 CT brain scan before, and after contrast injection showing
minimal enhancement of abscess wall and enhancing choroid plexus
A. Zakaria et al.

Discussion and review of the literature
Epidemiology
Nocardia species are strictly aerobic, Gram-positive, par-
tially acid fast, catalase and urase positive, non-motile
branching filamentous bacilli [3,10]. Edmond Nocard first
identified this bacterium in 1888 and 2 years later Eppinger
described the first patient with a cerebral abscess caused by
Nocardia asteroids. These microorganisms inhabit soil,
water, air and decaying vegetables [8]. Infection is acquired
primarily by inhalation from the environment and it spreads
from the lung via the blood to other sites especially skin
and brain [3,10]. It is a rare infection, the published
literature is limited and it consists largely of case reports or
small series [9,10]. In 1994 Mamelak et al. [10] reported
131 patients (including 11 of his own series) published in
the literature over 44 years (1950–1994). Central nervous
system Nocardiosis was reported in 2% of all cerebral
abscesses [8–10] and in 15–44% of patients with systemic
Fig. 5 CT brain scan, after 1 year, showing healthy bone flap and the
ventricular catheter
Fig. 4 CT brain scan showing hydrocephalus and residual abscess;
note the extent of the decompressive craniectomy
Fig. 3 CT brain scan after
deterioration showing massive
brain swelling, note obliteration
of the basal cisterns at the level
of the tentorial hiatus (a) and
foramen magnum (b)
Nocardia brain abscess

Nocardial infection [2,10]. Although up to two thirds of
Nocardia species infections occur in immuno-competent
individuals, more than half the patients with CNS disease
are immuno-compromised [3,9]. In Mamelak’s review, the
patient’s age ranged from 3 months to 85 years (mean age,
42±15.1 SD years). The male/female ratio was 2.3:1, and
34% of the patients were immuno-compromised [10].
Clinical presentation
Nocardia CNS infection may present as meningitis, diffuse
cerebral infiltration without localisation, granuloma with
giant cells or cerebral abscess/es [4,9]. The abscesses are
supratentorial in 57% of patients and infratentorial in 11%.
The lesions are single in 54%, multiple in 38% and of
unknown number in 8% of patients [10]. According to one
report, the principal presenting symptoms were focal deficit
in 42%, non-focal findings in 28% and seizures in 30% of
patients [10]. Our patient developed severe CNS infection
in the form of cerebritis, brain abscess, and ventriculitis.
The course of the disease was very serious and the clinical
condition of the patient rapidly deteriorated from GCS of
13/15 to decerebrate posturing and fixed pupils (GCS 4/15)
in 4 days despite aggressive treatment and surgical
drainage. The cause of deterioration in our patient was
ventriculitis and severe brain oedema.
Medical treatment
The antibiotic of choice is the synergetic combination of
trimethoprim/sulfamethoxazole. The recommended daily
intravenous dose of trimethoprim and sulfamethoxazole in
adults are 15–20 and 75–100 mg/kg respectively, while in
children the dose is 20 and 100 μg/kg respectively [1,5,
10]. If the patient did not respond to this first line treatment
or develops serious drug side effects (e.g. allergic reactions,
leucopenia or pancreatitis), second line antibiotics can be
used and consist of minocycline, imipenem, amikacin,
cefotaxime, metronidazole/flucloxacillin or combination of
these drugs [8,10]. Whatever regimen is used, it should be
continued for a period not less than 6 weeks via the
intravenous route followed by oral administration for a long
period and up to 1 year, in immuno-compromised patients
[10]. In our patient, the initial treatment (based on Gram
stain) consisted of broad spectrum antibiotics (ceftriaxone
and metronidazole), anti-tuberculous and antifungal (fluco-
nazole) medications. This was changed to trimethoprim/
sulfamethoxazole after obtaining the final results of
bacterial culture and sensitivity. Later on, it was changed
to a combination of intravenous minocyclin and merope-
nem because the patient developed leucopenia. The
intravenous meropenem was continued for 1 year while
minocyclin was discontinued after 6 months.
Surgical management
The optimal management of Nocardial brain abscess
remains unclear. Mamelak et al. [10] analysed 131 patients
with this condition along with their own experience of 11
patients. They recommended an empirical trial of “sulfa”
drugs if extraneural Nocardial infection is documented in a
clinically stable or an immuno-competent patient with a brain
abscess less than 2 cm in diameter. If the patient’scondition
deteriorates or if the abscess does not decrease in 4 weeks,
aspiration of the lesion should be performed to confirm the
diagnosis and to decompress the lesion. All abscesses larger
than 2.5 cm should be aspirated regardless of the immune
status of the patient. Serial CT or MRI brain should be done
every 2 weeks or after any clinical deterioration. If the abscess
enlarges after 2 weeks of treatment or remains unchanged for
4 weeks despite antibiotics, craniotomy should be performed
to excise the abscess. On the other hand Lee et al. [9]treated
nine out of 11 patients in their series by aspiration alone
which was repeated whenever necessary (one patient had
repeated aspiration four times), one patient had excision of
the lesion (radiologically resembled a glioma), and another
patient had aspiration once followed by craniotomy and
resection because of clinical deterioration.
From the experience of our patient, we believe that
initial aspiration is an optimal initial therapy followed by
antibiotics and aspiration repeated if the lesion does not
regress with antibiotics. In immuno-compromised patients
the treatment should be aggressive and attention should be
paid to the possibility of rapid deterioration of the patient.
Our patient deteriorated rapidly, the GCS dropping from 13/
15 to 4/15, the pupils being dilated. This deterioration was
attributed to severe progressive brain swelling and oedema
rather than hydrocephalus (as seen on the CT scan, with
obliteration of CSF spaces at the basal cisterns at the level
of tentorial hiatus and foramen magnum without apparent
change in ventricle size). Hence, urgent bi-frontal decom-
pressive craniectomy was necessary to save the patient’slife.
The bi-frontal decompressive craniectomy immediately low-
ered the intracranial pressure to normal and alleviated the fatal
effects of brain herniation. This effect was observed clinically
(pupils became small and equal and pressure readings were
normalised) immediately after surgery. Excision of the abscess
was not necessary in our patient as the infection responded to
antibiotics and repeated aspiration. We also emphasise the
different complications that might develop during the course
of the disease e.g. ventriculitis and hydrocephalus which
required surgical intervention.
Outcome
Nocardiosis involving the CNS is a very serious infection
and its mortality could be as high as 90% in patients who
A. Zakaria et al.

present with cerebral abscess [8]. With the advent of CT
scanning, the mortality dropped to 50% and with the
emergence of new generations of antibiotics the mortality
rate has further dropped to 33% in patients with single
cerebral abscess. Despite these developments, the mortality
of CNS Nocardiosis is still high and reaches 66% in
patients with multiple lesions [8,9]. Mamelak et al. [10],
reported that mortality varies with the type of management,
being 24% in patients treated with craniotomy and excision,
and raised to 50% in patients treated with aspiration alone.
For patients treated with antimicrobial drugs alone without
surgery the mortality was 30%, with 22% of CNS
Nocardiosis being diagnosed on autopsy.
Despite the fact that our patient was immuno-compromised,
the brain abscess was complicated with ventriculitis and
massive brain swelling, and the poor neurological status, he
survived and recovered to partial dependence (according to
Glasgow outcome score) after a year and continues to improve.
Conclusion
Atypical CNS infections should be considered in an
immuno-compromised patient. The course of Nocardia
brain abscess is severe and the patient’s condition might
deteriorate suddenly. The mortality in patients with Nocar-
dial abscess is three times higher than in patients with brain
abscesses caused by other bacteria. The initial treatment by
aspiration/drainage followed by trimethoprim and sulfame-
thoxazole is usually adequate. Bi-frontal decompressive
craniectomy can be a life saving procedure, and should be
considered if the patient develops severe brain swelling.
References
1. Baikie AG, MacDonald CB, Mundy GR (1970) Systemic
nocardiosis treated with trimethoprim and sulfamethoxazole.
Lancet 2:261 (letter)
2. Barnicoat MJ, Wierzbicki AS, Norman PM (1989) Cerebral
nocardiosis in immuno-suppressed patients: five cases. Q J Med
72:689–691
3. Beaman BL, Beaman L (1994) Nocardia species host–parasite
relationship. Clin Microbiol Rev 7:213–264
4. Beaman BL, Burnside J, Edwards B, Causey W (1976) Nocardial
infections in the United States 1972–1974. J Infectious Dis
134:286–290
5. Curry WA (1980) Human nocardiosis. A clinical review with
selected case reports. Arch Intern Med 140:818–827
6. Elwatidy S (2005) Bi-frontal decompressive craniotomy in a
6-month-old infant with post-traumatic refractory intracranial
hypertension. Pediatr Neurosurg 41:151–154
7. Elwatidy S (2006) Bi-frontal decompressive craniotomy for
malignant brain oedema. Neurosciences 11(4):241–247
8. Fleetwood IG, Embil JM, Ross IB (2000) Nocardia asteroides
brain abscess in immuno-competent hosts: report of 3 cases and
review of surgical recommendations. Surg Neurol 53(6):605–610
9. Lee GY, Daniel RT, Brophy BP, Reilly PL (2002) Surgical
treatment of nocardial brain abscess. Neurosurgery 51(3):668–672
10. Mamelak AN, Obana WG, Flaherty JF, Rosenblum ML (1994)
Nocardial brain abscess: treatment strategies and factors influencing
outcome. Neurosurgery 35(4):622–628
Comments
This is a single case report of a patient with a cerebral Nocardia ab-
scess complicated by ventriculitis. The infection was treated by anti-
biotics and aspiration of the abscess. Aggressive treatment of brain
swelling and subsequent hydrocephalus lead to recovery. The report
illustrates some of the hazards of this serious infection, occurring most
often in immunocompromised patients, and the steps to successful
treatment.
There is a good review of views about the place or timing of aspiration
versus excision of abscesses.
The authors state that the patient in this report recovered with "mod-
erate disability "according to the GOS. The details given of his state at
12 months surely place him in the "severe disability" category.
Peter Reilly
University of Adelaide
This is a detailed article on Nocardia brain abscess which covers the
subject almost fully. However the primordial question that may be asked
is, “Had the surgeon operated on the patient primarily and removed the
abscess, without rupturing it, rather than aspirating it, would that have
averted all the complications that ensued?”In fact the patient was later
subjected to a more serious surgery while he was in a much more critical
situation.The authors should be commended on their accurate discussion.
Professor Fuad S Haddad
American University of Beirut
Nocardia brain abscess