Keywords
awake craniotomy - cortical mapping - electrophysiological monitoring - language testing
- brain tumor
Introduction
“Arise! Awake! And stop not until the goal is reached”
– Swami Vivekananda
Awake craniotomy (AC) is a type of surgical procedure performed under local anesthesia
while keeping the patient intentionally conscious and alert during some portion or
the whole surgical procedure in order to save higher eloquent functions of the cerebral
hemispheres. Patients are only sedated while being awake during the basic part of
the surgery, thereby circumventing some of the complications often associated with
general anesthesia (GA) and endotracheal (ET) intubation, and therefore need for ventilators,
arterial lines, urethral catheters, and staying in intensive care units (ICU) postoperatively.[1] Furthermore, it is very helpful in avoiding morbidity and the side effects of GA.
It is especially beneficial in less-resourced, limited financial settings in low-
and middle-income countries (LMIC) as it also reduces the length of stay in the ICU.
Evolutionary Development of AC: A Historical Note
The earliest awake brain procedures were applied to the management of epilepsy in
ancient times, as evidenced by diverse archeological data from Egypt, where ancient
doctors attempted to do a trepanation of the skull, although it is unclear whether
drugs that served as GA were used at the time.[2] Robert Bartholow's experimental operation in 1874 trying to map functional areas
of the human brain in an awake patient has led to further investigations. Wilder Penfield
was one of the first scientists to strive it for intractable seizures in 1920, and
he and Andre Pasquet later published a paper on the use of intermittent anesthesia
and surgical peculiarities of AC. However, Victor Horsley from England was the first
to perform AC in 1886 by employing specific electric stimulation to localize the epileptic
focus. In terms of brain tumor surgery, Archer was the first surgeon to use AC in
1988. H. Duffau later emphasized the importance of preserving not only the cortical
centers but also the axonal pathways that connect the speech centers to the motor
areas and other parts of the cortex.[3] In 2003, several authors from Japan, India, and Thailand[4] reported one of the first cases of AC in Asia. Later, in 2007, some Malaysian authors
used AC for deep brain stimulation (DBS) surgery in Parkinson's disease.[4]
[5]
[6]
Present State of the AC Protocols in Use in Developed Countries
For AC, there are three main anesthesia protocols that are widely used (asleep-awake-asleep, asleep-awake-awake, and awake-awake-awake).[7] The most common technique is monitored conscious sedation, with the other option
being asleep-awake-asleep. Predominantly, the established protocols for surgical procedure
in most neurosurgical centers from developed countries include rigid head fixation
(Mayfield, Sugita, or other head clamps), use of brain neuronavigation system in order
to provide accurate skull positioning, and identifying the best angle to attack the
lesion.[8] Intravenous (IV) analgesia and local anesthesia—scalp block—are standard procedures
to deliver adequate analgesia. Japanese guidelines for AC recommend using the Wada
test to determine the dominant hemisphere stating that despite the availability of
noninvasive test like functional magnetic resonance imaging (fMRI) and magnetoencephalography
(MEG), they cannot absolutely determine the dominant side, because of the possibility
of pseudolocalization.[9] The goals of anesthetic management are to maintain general homeostasis, reduce interference
between anesthetic drugs, and improve the accuracy of electrophysiological recordings.
Key role is given to short-acting myorelaxants like rocuronium (1 mg/kg/min) and short-acting
opioid analgetic like remifentanil (0.05 μg/kg/min in 5 minutes). Hypnotic medication
is usually IV propofol (2 μg/kg/min). As a rule, laryngeal mask airway (LMA) or ET
tube should be used to secure the airway. Usually, later stage of anesthesia is maintained
with IV propofol and remifentanil infusions until awakening. For the awake part of
the operation, patients are given dexmedetomidine infusions starting at 0.2 µg/kg/h
with the dose titrated up to 0.7 μg/kg/h, as anxiolytic agent.[8]
Is There Hinderances to AC in Low-Resource Countries?
According to the investigations of Mofatteh Mohammad et al where the group of authors
have conducted a scoping review under PRISMA[20] (Preferred Reporting Items for Systematic Reviews and Meta-Analysis) - Scoping Review
of 19 studies (396 patients) utilizing guidelines from three databases (PubMed, Scopus,
and Web of Science) deficiency of appropriate infrastructure; deficiency of neurosurgeons,
nurses, and anesthesia personnel; long waiting time; and low quality of appropriate
training are some of the hindering factors of neurosurgical care in LMICs. The safety
and feasibility of AC in low-resource settings where modern and expensive technologies,
such as fMRI and intraoperative MRI (iMRI), intraoperative cortical mapping, and electrophysiology,
are not available are a serious challenge. Nine studies (47.4%)[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19] reported infrastructure limitations as an obstacle to performing AC operations with
one study noting that lack of head pins at their hospital served as an obstacle to
AC.[20] In our opinion, despite these constraints in technical and equipment facilities,
awake surgery can be executed in LMIC settings. We will share our experience in a
single institution study.
Aim
The purpose of this study is to concisely describe the simplified protocol of awake
brain surgeries that can be implemented in a limited financial setting in LMICs and
to share our first experience.
Methods
The study was done at the Republican Scientific Practical Medical Center of Neurosurgery
(Tashkent) in 2022 to 2023, including 25 patients with the tumor of the left frontotemporal
lobe, involving Broca's and Wernicke's areas.
Patient Selection
All patients were selected according to the criteria accepted by most awake surgery
protocols: patients with no speech or sensorimotor deficits, patients with mild disturbances
that are tolerable in general speech assessment, patients who are able to understand
spoken speech, and patients with had healthy airways. Pre- and postoperative functional
outcomes were assessed via a neurological examination.
Preoperative Preparation
All 23 patients were preoperatively prepared psychologically and physiologically by
the operating team (assistant neurosurgeons), basically by showing the surgical videos
from previous cases, by showing pictures, flash cards, and verb generating ability
and semantic function. No special psychiatrists or psychologists were involved. All
the patients received carbamazepine (400 mg/d) preoperatively in order to avoid intra-/postoperative
seizures. Some institutions preferred using bolus IV fosphenytoin (15 mg/kg) as seizure
prophylaxis.[21] But we realized that oral prescription of carbamazepine is also effective enough.
We have used MRI tractography to evaluate white matter tracts and speech pathways
as needed.
Anesthesiologic Features
Although key role is given to short-acting myorelaxants like rocuronium (1 mg/kg)
and short-acting opioid analgesics like remifentanil (0.05 mg/kg in 5 min), due to
unavailability of identical alternatives in our country, we used fentanyl titrating
its dose. As a sedative medication used IV propofol (2 mg/kg/1). A laryngeal mask
device (LMA) is a useful tool to secure the airway. Later stage of anesthesia was
maintained with IV propofol and fentanyl infusions until awakening. For the awake
part of the operation, patients were given dexmedetomidine infusions starting at 0.2
µg/kg/h with the dose titrated up 0.8 µg/kg/h, as anxiolytic agent.
As for local anesthesia providing scalp block, six nerves of the scalp on the ipsilateral
side were blocked using the solution of bupivacaine 0.5% (30 mL) with epinephrine
(1:200,000). (nn. supraorbitalis, supratrochlearis, auriculotemporalis, zygomaticotemporalis,
nn.occipitalis major et minor).
Setting up OR at a Minimum Basis
Positioning of the patients: For all of our patients, we used the right-sided lateral decubitus or the lateral
park bench position, which we found more convenient for patient communication and
airway management ([Fig. 1]). We have used a single drape to cover the patient's head and body leaving open
access to the facial side for communication. As we do not have Mayfield or other type
of rigid head fixating clamps, we used a simple horseshoe head rest or head holder.
In our opinion, it was even better avoiding the three-point head rigid fixation, as
it simplifies airway management for the anesthesiologist during the intubation/extubation
phase in the middle of the surgery and creates more comfort for the patient, decreasing
the range of anxiety and painful uneasiness after awakening.
Fig. 1 (A) Operating room setup and patient's position. (B) The patient is awakened, and the language mapping is started.
Cortical/Subcortical Stimulation and Language Mapping
Neurophysiological brain stimulation in the awake stage of the surgery was conducted
using the Inomed Intraoperative Neuromonitoring (IONM) system (Tuttlingen AG, Germany).
Basically, stimulation was applied with a bipolar stimulation frequency of 60 Hz,
pulse duration of 1 millisecond, and intensity ranging from 3 to 6 mA by the method
of Ojemann and Mateer.[22] We begin our cortical stimulation from 3 mA then gradually elevating until we identify
positive mapping sign- aphasia/dysphasia, if neurophysiologist detect after discharge
potentials, we stop increasing current strength. In our experience, usually maximum
amperage needed to stimulate cortex was 6 mA. And no seizure occurred. Positive mapping
zones were marked with square sheets of paper. The loci around the area of interest
showing no clinical signs of dysphasia are marked as negative zones, and corticectomy
can be performed safely in these areas. There was no significant mass effect on the
sensorimotor tracts or related clinical presentations, so we decided to map only the
language cortex. No speech therapists or neuropsychologists were involved in our surgeries
since language functions were tested by a third assistant neurosurgeon standing nearby,
using the standard speech test to evaluate object naming, verb generation, calculus,
semantic association, etc.
Intraoperative Neuropsychological Tasks
All the patients underwent preoperative and intraoperative neuropsychological assessment
tests. We recorded the results of the tasks before and after stimulation of the left
frontotemporal area where the tumor was located. A task is considered to be successfully
completed if the patient answers correctly to the task questions and “failed” if he
or she answers wrongly or shows symptoms of clear dysphasia or fails to the answer,
which is denominated as “speech arrest.” Intraoperative tasks included counting numbers
from 1 to 10 and reverse counting from 10 to 1, object naming test (household objects
and pets), pyramid and palm tree test (PPTT) for a semantic association test, word
repetition test to check transcortical speech, and simultaneous left leg movement
to check multitasking capability. During resection of the tumor, the arcuate fasciculus
was stimulated to check for phonological disorder and mental activity, and the superior
longitudinal fasciculus was stimulated to register working memory and attention using
mono- and bipolar stimulating profiles.
Results
Tumor Resection
In our study, AC was successfully performed in 23 of 25 patients, achieving near total
resection in 16 (69.5%) patients, subtotal resection in 4 patients (17.39%), and partial
resection in 3 (13.04%) patients. Speech test was not possible in two patients due
psychological instability—agitation and fear—during the awake phase, so they were
reintubated with GA. There was no mortality in early or postoperative period.
Twenty-three cases in our study, tumors resection in the left dominant hemisphere
was performed using blunt dissection, bipolar coagulation, and the Cavitron Ultrasonic
Surgical Aspirator (CUSA) ultrasonic aspirator (Integra). The extent of tumor resection
confirmed was with postoperative computed tomography (CT). Sixteen (69.5%) patients
had near total resection, 3 patients (13.04%) had partial resection, and 4 patients
(17.39%) had subtotal resection. There was no mortality in the early or postoperative
period. After language mapping was complete, we reintubated our patients with propofol.
Illustrative Case
We present our first case of awake brain tumor surgery in Uzbekistan. A 55-year-old
woman was admitted to our center with headache, slight memory impairment, and general
fatigue. The symptoms began 4 months ago, and she had previously seen an oncologist
and empirically received 12 days of chemotherapy with temozolomide on the spot. (Unfortunately,
they have lost the first MRI series, so we could not present it.) The patient rejected
surgical excision and opted for chemotherapy ([Fig. 2A]). After 1 month of chemotherapy with temozolomide, there was slight improvements
in her neurological status, and her headaches regressed. However, after 4 months,
she experienced a severe headache, followed by memory deficit. She underwent brain
MRI, which revealed an enlargement of the tumor's margins. She contacted us immediately.
She was neurologically examined. She was right-handed and had a slight amnestic disorder.
There were no motor or sensory deficits in the limbs.
Fig. 2 (A) First magnetic resonance imaging (MRI) after having chemotherapy with temozolomide.
Axial and sagittal MRI sequences on fast spoiled gradient echo (FSPGR) detected a
17 × 14 mm hyperintense lesion on the posterior parts of the inferior frontal gyrus
of the frontotemporal area of the left hemisphere. (B) Preoperative MRI from May 29, 2021, showing progressive tumor growth and marked
pertitumorous edema causing a brain shift of 9.6 mm. MRI showing active growth of
the tumor with large cystic component, causing midline shift and mass effect. (C) Early postoperative computed tomography (CT) depicts remnants of the tumor in the
peri-insular area and slight brain edema of the tumor bed.
On May 29, 2021 at the time of admission to our hospital, the volume of the tumor
was significantly increased, with signs of hemorrhage within the tumor, causing serious
mass effect and peritumoral edema apparent on MRI slices ([Fig. 2B]) done the day after her general condition worsened with the symptoms of strong intracranial
hypertension and slight amnestic disorders. After discussing the case (her neurological
status and present speech abilities) with our team, we decided to utilize the chance
of awake brain tumor surgery. Although it was a new procedure for everybody, we prepared
all the necessities. The patient was trained by our team undergoing speech tests to
control expressive and impressive speech components. There was some nominal dysphasia
(about 12% wrong answers), which was in our opinion quite tolerable. Nevertheless,
she felt comfortable and confident during the awake stage of the operation. The OR
was preoperatively prepared; the team consisted of neurosurgeons, a neurophysiologist,
and an anesthesiologist. Everybody preoperatively trained and learned their function
during the upcoming procedure.
Anesthesia
Anesthesia was done utilizing “asleep-awake-asleep” protocol for our awake brain tumor
surgery as planned. During the induction phase of anesthesia, brain relaxation was
provided using an IV infusion of mannitol 1 mg/kg and hyperventilation when needed.
As an analgetic, fentanyl was used, and dexmedetomidine was injected 5 minutes prior
to the awakening phase.
Surgical Procedure
We used a park bench position, softly fixing the head on a horseshoe head rest using
tapes, elevating the head side to 30 degrees. Skin electrodes were placed appropriately.
After local anesthesia with bupivacaine 0.5% - 30ml with epinephrine (1:200,000),
the skin was incised in a regular curvilinear fashion, performing a regular left-sided
frontotemporal craniotomy. The dura was opened in a usual horseshoe fashion and wrapped
with lidocaine-soaked cottonoids. The brain was relaxed. The anesthesiologist awakened
the patient smoothly as planned. After regaining adequate and clear consciousness,
our neurologist started his conversation with the patient. The answers to the given
tests were up to 90% correct and she felt no panic or pain. Afterward, we started
electrical stimulation of the cortex with a bipolar stimulator, giving a current of
3 mA. As we have not detected speech arrest or signs of disarticulation, we proceeded
with 5 mA. Finally, we elicited a region of the inferior frontal gyrus where she had
experienced “speech arrest.” After removing the stimulator, her abilities were regained,
and we marked this point with sterile material (square-shaped paper). The same procedure
was performed to reveal the sensory speech area of Wernicke. As soon as there was symptom of semantic aphasia and paraphasia in the subcortical
level, we ceased resection of the tumor to avoid damaging the subcortical tracts and
the arcuate fasciculus. The tumor was removed in a piecemeal fashion, subtotally.
During resection of the deeper parts of the tumor in the vicinity of the arcuate fasciculus,
subcortical stimulation was conducted with a monopolar stimulator (with 4–5 mA). Some
remnants of the tumor were left to preserve the deep arcuate fibers. The approximate
extent of the tumor resection was 80%. After removing the tumor, there was an obvious
slight congestion of the sylvan veins, most probably because of the use of retractors.
Pathology report was anaplastic astrocytoma G III.
Postoperative Period
During the postoperative period, the first day was uneventful. She was talking, with
slight errors in naming objects (amnestic aphasia). But later, in the morning of the
second day, nurses reported that she was not quite alert and responsive. Neurologically,
she was somnolent (Glasgow Coma Scale [GCS]: 9–10). CT scan showed a huge edema and
brain shift ([Fig. 2C]), which pushed us to perform an emergency decompressive craniotomy. The patient
awakened successfully after 5 to 6 hours and the postoperative period was uneventful.
Postoperative CT shows ([Fig. 3A,B]) venous edema bony decompression. Fortunately, her speech performance did not change
in the postoperative period. She was discharged on postoperative day 10 with a recommendation
to receive adjuvant chemotherapy and radiotherapy. Later, she received the necessary
chemotherapy and radiotherapy and came to us with a follow-up MRI after 5 and 11 months
showing a very good effect of adjuvant treatment ([Figs. 4] and [5]). Her amnesia and speech disorders were significantly improved.
Fig. 3 (A,B) Computed tomography (CT) after performing decompressive craniectomy of the left
frontotemporal region. Hyperdense areas of the left frontotemporal lobes show persisting
ischemic changes.
Fig. 4 Follow-up magnetic resonance imaging (MRI) investigation 5 months later after adjuvant
chemotherapy depicts a small remnant of the tumor, which was controllable with chemotherapeutic
agents.
Fig. 5 Follow-up magnetic resonance imaging (MRI) 11 months later (April 14, 2022) after
adjuvant chemotherapy. Axial T1- and T2-weighted images show the remnant of the tumor
has decreased significantly.
Discussion
Utilization of awake surgery provides preservation of the eloquent cortical centers
and subcortical white matter tracts. Direct electrical stimulation of the eloquent
cortex serves as a mapping of language, motor, sensory, and other essential higher
brain functions like emotional processing, which is localized in the insular area.[23] Using intraoperative motor and sensory evoked potential can help find out the precise
localization of the eloquent brain areas.[24]
Surgical outcomes in such patients are directly related to the quality of tumor excision
using both advanced preoperative and intraoperative technologies such as diffusion
tensor imaging (DTI), iMRI, and navigation systems. Some studies suggest co-registration
of iMRI with preoperative fractional anisotropy color maps is useful for intraoperative
localization of the subcortical tracts.[25] These tools along with intraoperative cortical mapping and language testing can
guarantee better surgical outcomes and quality of life.
The main goal of awake surgery is to achieve a safer and wider resection by preserving
the eloquent areas of the brain. However, the extent of resection truly correlates
with longer progression-free survival rates of patients with grade 2 and 3 gliomas
except for aggressive isocitrate dehydrogenase (IDH) wild type of diffuse astrocytoma
and anaplastic astrocytoma.[26] Taylor et al highlighted the usefulness of AC even as a routine adjunct procedure
not only for eloquent cortex but also for all supratentorial locations regardless
of the functionality of the cerebral cortex.[27] In addition to this, the effectiveness of awake surgery can be improved by employing
iMRI[28]
[29] and navigation.[30] For sure, having all high-tech facilities indeed intensifies the quality and extent
of resection of tumors in eloquent brain areas. But in our opinion, there are still
some adjunct equipment that can be partially abandoned in order to adjust awake surgery
protocol in LMIC on a minimum basis, in less equipped conditions without losing the
safety and basic philosophy of the procedure.
Comparing the guidelines and protocols provided by several authors, according to the
investigated literature, a small number of the patients still complain of pain and
anxiety in the postoperative period, which is mostly attributed to the rigid pinning
of the skull to a Mayfield head pin.[8]
[31]
[32]
[33] Almost all awake surgery protocols recommend the use of rigid head fixation. It
provides a stable and still head position, thereby allowing the surgeon to work safely
and confidently. At the stage of awakening, many patients feel uncomfortable because
their skull is fixed with pins and they unable to move which is a key factor for anxiety.
We consider to alleviate this inconvenience and used soft fixation with tapes. In
contrast to that, we have used a simple horseshoe head holder, which has many benefits.
It serves as a gentle cushion for patient's head, and with its use, airways and laryngeal
mask replacement are easily accessed, causing no pain. To protect and avoid undeliberate
head movement during awake phase, we use simple medical tapes to softly fix the patients
head, which are applied loosely and softly. In our cases, due to an absence of iMRI,
navigation system, and preoperative DTI studies, our main goal was a bit limited.
We performed an awake surgery for the first time in the history of Uzbekistan, and
we initially decided to preserve the language function. These inconveniences with
pre- and intraoperative assessment along with intraoperative language and movement
testing limited our success in terms of maximum resection in order to avoid damaging
the descending motor pathways in the corona radiata, which might be closer to or within
the tumor. This resulted in a shortening of tumor progression and the patient experienced
further tumor growth even after chemotherapy and radiation therapy. Even a recent
report[30] suggested that awake surgery might not be realized in developing countries for the
next several years. However, our experience shows that this procedure can be utilized
in these conditions as well. The example of an awake surgery in our report proves
that procedures with limited technologies can be helpful for patients with gliomas
in the eloquent areas of the brain since it at least could help preserve the language
and movement function by achieving a safer maximum resection compared to conventional
brain tumor surgery.
Summarizing the above-reviewed literature and our team's experience in this direction,
we highlights some points in this regard as following section.
Conclusion
High-grade gliomas within the functional areas of the brain usually penetrate into
the critical subcortical tracts that are responsible for complex high brain functions.
These tumors often represent multiple symptoms such as language and movement problems,
especially involving the eloquent brain areas. While patients with severe preoperative
speech disturbances cannot qualify for awake surgery, patients with mild or asymptomatic
patients are candidates for awake surgery. Although, LMICs do not have these high-tech
facilities, we should not wait for it. We need to improve our current state and go
further to reach our goals and alternative ways. Intraoperative cortical mapping and
language testing with awake surgery can be affordable, and even employing them without
the above-mentioned tools can lead to better outcomes in terms of preservation of
language and movements compared to conventional brain tumor surgical procedures.
Simplifying the Technique in the Setting of Limited Financial Conditions
Simplifying the Technique in the Setting of Limited Financial Conditions
-
Performing awake brain surgery is necessary and feasible even in limited financial
settings.
-
There is no need for a rigid head fixation during surgical procedures since adequate
scalp block with long-acting anesthetic agent provides smooth and comfortable procedure
while also eliminating any possible range of pain and discomfort for the patient.
In our case series, we have used only medical tape to provide gentle and soft fixation.
-
There is no need for remifentanil, since timely titration of an ordinary fentanyl
can sustain pre-awake anesthesia if well calculated by the anesthesiologist.
-
There is no need for neuronavigation. Good knowledge of craniometric points and intraoperative
cortical anatomy along with qualified mapping technique gives enough information at
a real time (excluding deep-seated lesions).
-
There is no need for intraoperative electroencephalographic (EEG) monitoring, since
neurological evaluation of subtle signs of awakening helps in control of (increasing/decreasing)
anesthetic agents.
