Key-words:
Brain abscess - cerebral infarction - decompressive craniectomy
Introduction
Stroke is the second leading cause of mortality in the world after coronary artery
disease. The global burden of stroke has risen substantially in the past two decades,
especially in low and middle-income countries.[[1]] The outcome of malignant stroke in general is poor. Due to the recent trend of
decompressive craniectomy, there has been an increase in the number of survivors at
the cost of higher morbidity.[[2]] There have been many complications described as a sequela to stroke. Brain abscess
is a rare complication which is less described among survivors of stroke. The usual
presentation of brain abscess following stroke is due to symptoms of raised intracranial
pressure. In this report, we describe two cases of malignant stroke in young patients
who survived with decompressive hemicraniectomy. They were found to have a large brain
abscess at the site of the previously infarcted region, which was detected incidentally
at the time of cranioplasty. We discuss the management strategies along with a brief
review of the literature of brain abscess following decompressive craniectomy for
stroke.
Case Reports
Case 1
A 35-year-old male presented to the emergency department with a history of sudden
onset of headache and vomiting. His symptoms progressed the next day during which
he noticed weakness of right side of the body and difficulty in speech. He was referred
to our tertiary unit after 2 days of illness. His Glasgow Coma Scale (GCS) at admission
was E4V1M5 with Grade 0 power in the right upper and lower limb along with brisk reflexes
and a positive Babinski's sign. The National Institutes of Health Stroke Scale score
was 25. An emergency computed tomography (CT) brain was done, which showed mild hypodensity
in the region of the left middle cerebral artery territory. His routine investigations
revealed that he had diabetes mellitus and also high packed cell volume values, possibly
due to dehydration. A diagnosis of middle cerebral artery infarct was made after which
he was started on heparin infusion. After 1 day, he deteriorated in his consciousness
with a GCS of E2V1M5 with mild pupillary asymmetry. A repeat CT scan showed that there
was a large infarct in the area supplied by the left middle cerebral artery with a
midline shift of 7 mm [[Figure 1]]a. A decision to do a decompressive hemicraniectomy was made at this point. A large
fronto-temporo-parietal flap was raised, and the underlying bone was removed during
the surgery. The dura was opened. We do not breach the pia for decompressive hemicraniectomy
for stroke. A sizeable synthetic dural graft, G-Patch (Surgiwear Inc.) was placed
over the brain, and skin was closed in layers. The resected large piece of the bone
flap was placed in the subcutaneous pouch created in the abdomen.
Figure 1: Computed tomography brain plain. (a) Large hypodensity in the left middle cerebral
artery territory with mass effect on the lateral ventricle and midline shift to the
right. (b) Postoperative image after decompressive craniectomy showing adequate decompression
of the left cerebral hemisphere
Postoperative period and follow-up
The patient was mechanically ventilated and gradually extubated after 2 days. His
consciousness improved to preoperative status. Postoperative CT showed an excellent
decompression of the infarcted area and adequate room for the edematous brain [[Figure 1]]b. He had a moderate-to-severe chest infection complicating his recovery period
for which he was treated with appropriate antibiotics. During the same period, he
developed gradual swelling of his right lower limb. A Doppler study revealed deep
vein thrombosis involving common femoral and popliteal veins. He was continued on
heparin infusion along with warfarin. He recovered well over 1 month and at discharge
with advice to continue with warfarin and insulin. At discharge, he was E4VAM5 with
right hemiplegia.
Readmission and surgery
He reported to us for cranioplasty after 6 months. There was no history of fever during
any time of follow-up. He was E4VAM5 with Grade 1 power in his right upper and lower
limb. According to bystanders, he had improved well over the previous 6 months. A
CT brain was taken before cranioplasty, which showed a large hypodense ring-enhancing
lesion in the region of previous infarct [[Figure 2]]a and [[Figure 2]]b. A diagnosis of possible brain abscess was arrived at, and cranioplasty was deferred.
All routine blood investigations including culture were sent to detect any signs of
infection, but they were normal. His blood sugar values were within control as he
was on insulin. He underwent open surgery for the draining of brain abscess. After
skin incision the synthetic dura was excised completely. The abscess cavity was opened,
and around 20 ml of pus was let out and send for cultures and bacteriology. The capsule
of the brain abscess was excised as much as possible leaving a part of it on the medial
side adjacent to the ventricle. This was done as we did not want to contaminate the
ventricles. The abscess cavity was washed with gentamicin solution, and the brain
was covered with a large fascia lata graft harvested from the left thigh and skin
closed in layers.
Figure 2: (a and b) Computed tomography brain plain showing ring-enhancing lesions in the left
frontoparietal region with central hypodensity. Note the absence of any mass effect
in spite of the large size of the lesion
Follow-up after the second surgery
The patient recovered well from the operation. The results of the culture of pus,
as well as the dural prosthesis, were sterile. He was given intravenous imipenem and
vancomycin for 4 weeks with advice to continue oral antibiotics for 2 weeks at home.
He was advised to continue physiotherapy at home. He was on regular follow-up in the
outpatient department. At 6 months, his power in the limbs had improved to Grade 3.
At 1 year, CT scan was done, which showed resolution of abscess after which he underwent
cranioplasty with autologous bone flap [[Figure 3]]. He is on regular follow-up for 2 years, and there are no signs of raised intracranial
pressure or intracranial infections. He is fully conscious and has right hemiplegia
and aphasia.
Figure 3: Computed tomography brain plain - Postcranioplasty after 1 year showing no residual
abscess
Case 2
A 26-year-old male had a sudden onset of headache and altered sensorium. He was brought
into the emergency department and found to be delirious. On examination, he was febrile,
and GCS was 10/15. Both pupils were bilateral sluggishly reacting. He had right hemiplegia.
He was a known case of congenital heart disease, for which he had undergone surgery
in childhood. All blood investigations including culture were sent. He tested positive
for immunoglobulin M Brucella antibody suggesting brucellosis. However, blood culture
did not grow any organism including Brucella. He was also found to have concomitant
endocarditis. An emergent CT scan was taken, which showed a left-sided middle cerebral
artery infarct with mass effect and midline shift of 10 mm [[Figure 4]]a and [[Figure 4]]b. The need for emergency surgery was discussed with the bystanders, after which
he underwent left fronto-temporo-parietal decompressive craniectomy. The bone flap
was kept in the abdomen. Postoperatively, his consciousness improved. He recovered
well but was aphasic and was discharged after 2 weeks with a GCS of E4VAM5. After
2 weeks, he turned up for a follow-up. He was fully conscious, had expressive aphasia,
but was wheelchair-bound. He was asked to continue doxycycline for Brucella, as advised
by the general medicine department.
Figure 4: Computed tomography brain plain (a and b) - Acute infarct in the left frontal-temporal
and parietal region with mass effect on the lateral ventricles and midline shift toward
the right side
After 4 months, he was readmitted to the neurosurgery department for cranioplasty.
He had no worsening of consciousness or weakness of limbs during this period. He underwent
a routine preoperative CT scan, which showed a large loculated hypodense collection
in the infarcted area of the brain [[Figure 5]]a and [[Figure 5]]b. After arriving at a diagnosis of postdecompressive craniectomy brain abscess,
he underwent surgery for the excision of brain abscess. At surgery, the cavity was
aspirated, which yielded around 25 ml of pus. The abscess, along with the fibrous
capsule, was excised in toto. The pus which was sent for culture turned out to be
sterile. He recovered well after the surgery. The postoperative CT scan showed the
total excision of the abscess [[Figure 6]]a and [[Figure 6]]b. Two months after the excision of brain abscess, he was called up for cranioplasty.
He underwent a cranioplasty with the autologous bone flap 2 months after the second
surgery. He recovered well after cranioplasty, and a postoperative CT scan was taken,
which was uneventful [[Figure 6]]c and [[Figure 6]]d. At the last follow-up, he had partial recovery of his speech and was fully conscious
with a GCS of E4V4M6. A summary of both cases is detailed in [[Table 1]].
Figure 5: Computed tomography brain plain (a and b) - A well-defined oval-shaped hypodensity
(suggestive of brain abscess). Occupying the left fronto-temporo-parietal region.
The decompressive bone defect is also visualized
Figure 6: Computed tomography brain plain (a and b) - Postbrain abscess excision shows total
excision of the abscess. There is ex vacuo dilatation of the ventricular system on
the left side. (c and d) - Postcranioplasty computed tomography brain
Table 1: Summary of cases of brain abscess
Discussion
The occurrence of brain abscess following stroke was less well recognized until number
of case reports surfaced in recent years in this regard. A thorough search on popular
medical search engine PubMed and Google Scholar using the Boolean terms, “Stroke”
AND “Complications” AND “Brain abscess” yielded only 19 case reports where a similar
association was described.[[3]],[[4]],[[5]],[[6]],[[7]],[[8]],[[9]],[[10]],[[11]],[[12]],[[13]],[[14]],[[15]],[[16]],[[17]],[[18]],[[19]],[[20]],[[21]] Among them, four reports were excluded in this review because only abstracts were
available in English.[[3]],[[4]],[[14]],[[17]] Of the remaining 15 cases reported so far, only two patients had abscess following
a decompressive craniectomy for stroke.[[13]],[[18]] In both reports, the patients were symptomatic with respect to the secondary brain
abscess, but in our case both the patients were relatively asymptomatic. The condition
of both had very much improved over time from the initial surgery for stroke to the
development of an abscess. Our report hence emphasizes the role of routine CT in the
recovery period after decompressive craniectomy following ischemic stroke. This can
detect such silent abscesses and prompt early intervention. Following the universal
dictum that abscess anywhere in the body has to be drained, we decided that the best
option is to go ahead with surgery and drainage of the abscess. During surgery, we
had instilled gentamycin solution to wash the abscess cavity. This was done to decrease
the antibacterial load within the remaining abscess cavity. Such practices have been
recommended by many authors in the literature.[[22]],[[23]],[[24]] Furthermore, it was decided that the antibiotics needed to be stepped in view of
high incidences of organisms not being discovered in routine cultures in our experience.
Such sterile abscess form up to 40% of large series of brain abscess reported in the
literature.[[25]] Brain abscess after stroke has been seen to occur mostly in the elderly population,[[3]],[[4]],[[5]],[[6]],[[9]],[[10]],[[11]],[[12]],[[15]],[[18]],[[21]] with only a few cases reported in the younger age group.[[7]],[[8]],[[11]],[[18]] The combination of immunocompromised states such as surgery, diabetes, and advanced
age may have played a role in the development of brain abscess in the setting of stroke.
Although both our patients were young, the former was a diabetic. This, along with
surgery, might have been a predisposing factor for developing an intracranial infection
in him. Our second patient was a case of congenital heart disease and had Brucella
meningitis and endocarditis. Neurobrucellosis has known to cause cerebral infarction
[[26]],[[27]] as well as brain abscess.[[28]],[[29]],[[30]] In addition, it has been reported that the central nervous system itself is an
immunodepressive state after stroke which predisposes patients to various infection.[[31]],[[32]],[[33]] The pathogen responsible for abscess after stroke has been mostly due to Staphylococcus
spp. However, there have been instances where rare organisms [[5]],[[9]],[[18]] or negative cultures [[6]],[[11]],[[12]],[[18]] have been reported.
The time for the development of brain abscess following stroke varies in the literature
from as early as few weeks,[[6]],[[7]],[[8]],[[11]],[[12]],[[13]],[[18]] months,[[9]],[[15]],[[18]],[[21]] or even an year.[[10]] In our patients, the exact time for the development of abscess is not known since
imaging was not repeated after initial discharge to the time of readmission for cranioplasty.
Overall, the prognosis of the development of abscess following ischemic stroke is
poor. Death or recovery with neurological deficit predominates the literature review
of this condition. In the review by Ramesh et al. which included 19 cases of brain
abscess following stroke, there were five deaths and four partial recoveries of neurological
status.[[21]] The early surgical intervention followed by appropriate antibiotics will decrease
the mortality associated with this condition.
Conclusion
Brain abscess following DC for stroke is rare. The symptoms and signs are usually
masked by the accommodative nature of the decompressed brain. A routine CT scan should
be considered in all patients undergoing cranioplasty before the procedure. Surgical
excision followed by interval cranioplasty should be done in all cases.
Declaration of patient consent
The authors certify that they have obtained all appropriate patient consent forms.
In the form, the patients have given their consent for their images and other clinical
information to be reported in the journal. The patients understand that their names
and initials will not be published and due efforts will be made to conceal their identity,
but anonymity cannot be guaranteed.