C h a l l e n g e s i n Pe d i a t r i c

N e u ro a n e s t h e s i a
Awake Craniotomy, Intraoperative Magnetic
Resonance Imaging, and Interventional
Neuroradiology

Craig D. McClain, MD, MPH*, Mary Landrigan-Ossar, MD, PhD

KEYWORDS
 Pediatric  Intraoperative magnetic resonance imaging  Awake craniotomy
 Neurointerventional  Neuroendovascular  Pediatric neuroanesthesia

KEY POINTS
 There are many complexities to the care of children undergoing awake craniotomies.
 The anesthesiologist must be prepared to deal with a variety of urgent and emergent intra-
operative scenarios.
 When the techniques of cortical mapping are combined with an awake, responsive pa-
tient, optimal outcomes can be realized.
 Intraoperative magnetic resonance imaging offers high-resolution intraoperative images
that can assess the extent of resection in pseudoreal time.
 Angiography and embolization are frequent procedures performed in the neurointerven-
tional suite to address a variety of pediatric neurovascular lesions.

INTRODUCTION

Anesthesiologists involved in caring for children undergoing neurosurgical procedures

are required to have an intimate understanding of normal neurocognitive devel-
opment, the effects of anesthetics on the developing nervous system, the fundamental
differences between children and adults, and the implications of these surgical
approaches to children. Several surgical approaches such as image-guided proce-
dures and awake craniotomies add to the complex environment faced by the anesthe-
siologist. In addition, the neurointerventional suite has become increasingly used as

Disclosures: the authors have no disclosures to make.

Department of Anesthesiology, Perioperative and Pain Medicine, Boston Children’s Hospital,
Harvard Medical School, 300 Longwood Avenue, Bader 3, Boston, MA 02115, USA
* Corresponding author.
E-mail address: craig.mcclain@childrens.harvard.edu

Anesthesiology Clin 32 (2014) 83–100

http://dx.doi.org/10.1016/j.anclin.2013.10.009 anesthesiology.theclinics.com
1932-2275/14/$ – see front matter Ó 2014 Elsevier Inc. All rights reserved.
84 McClain & Landrigan-Ossar

children with a variety of neurovascular lesions present for often lengthy and compli-
cated procedures for definitive diagnosis or treatment. Planning and executing safe,
age-appropriate perioperative care in these environments is challenging. This article
offers some insight into the complexities of care of children undergoing awake crani-
otomies as well as procedures in intraoperative magnetic resonance imaging (iMRI)
suites and neurointerventional radiology.

AWAKE CRANIOTOMY
History of Awake Craniotomy
Evidence of craniotomy predates the invention of surgical anesthesia by several
millennia. There is evidence of trepanation (creating a hole through the skull and
dura) in human skulls unearthed in France from approximately 6500 BC.1 In addition,
it is clear that several pre-Columbian societies in Mesoamerica practiced trepanation,
most notably, the Incas.2,3 During the Middle Ages and Renaissance in western
Europe, trepanation was performed to alleviate headaches and seizures.4 Dutch
painter Hieronymus Bosch famously captured this practice in his painting, The Extrac-
tion of the Stone of Madness, from the late fifteenth century.
The modern use of awake craniotomy (AC) began in the second half of the nineteenth
century, when local anesthetics became widely available. With good local anesthesia,
Horsley was able to perform ACs.5 However, the modern understanding of the benefit
of AC began in 1951, when Wilder Penfield, the first director of the famous Montreal
Neurologic Institute, published his landmark monograph, Epilepsy and the Functional
Anatomy of the Human Brain.6 Penfield described the use of craniotomy performed
under local anesthesia only to facilitate resection of epileptogenic foci. Before resection,
Penfield stimulated various locations of the cortex and observed the responses in the
awake patient. This practice allowed him to generate cortical maps of motor and sensory
areas, which result in cortical homunculus.
The 1960s brought the advent of neuroleptic anesthetic techniques, which continued
to provide a responsive patient but offered some degree of analgesia and sedation in
order to tolerate prolonged awkward positions.7 A combination of drugs such as dro-
peridol and fentanyl were commonly used to facilitate a patient who was drowsy and
comfortable, yet still able to arouse to stimulation and follow commands. The downside
of prolonged use of dopaminergic drugs became apparent when the occurrence of side
effects, including extrapyramidal effects and dysphoria, was noted.
AC as a method of treating seizure foci and tumors became popular again in the
1990s and early 2000s, with the widespread use of shorter-acting hypnotic agents
and opioids, such as propofol and remifentanil.8,9 Current anesthetic techniques
use a wide range of agents. Dexmedetomidine, an a2 agonist with sedating and anal-
gesic properties, is a relatively newer drug that offers some distinct advantages over
other techniques using sedative hypnotics and opioids.10,11 One of the most advanta-
geous aspects of dexmedetomidine is its ability to offer mild analgesia and good seda-
tion without compromising the airway.

AC in Children
Equipment
No special equipment is needed for the performance of AC. The anesthesiologist
should have the operating room (OR) prepared for the same problems that may be
encountered during a craniotomy under general anesthesia. Invasive blood pressure
monitoring is useful. The anesthesiologist should be prepared to convert to a general
anesthetic if needed. Airway management while the patient is in head pins can be
 Challenges in Pediatric Neuroanesthesia 85

particularly problematic. It can be lifesaving to have properly sized supraglottic airway

devices within easy reach. More advanced airway tools may be necessary to provide
definitive airway management.
The anesthesiologist must be prepared to deal with a variety of urgent and emergent
intraoperative scenarios. This strategy applies to any surgical case, but may be more
important for an AC. Seizures, which can occur during cortical mapping in an AC, must
be quickly and effectively controlled to prevent patient injury. Seizures are more
common in younger patients with frontal lobe lesions.12 Generalized tonic clonic
movements can be devastating to a patient who is fixed in head pins. Injuries that
may result can range from scalp lacerations to skull fractures and even cervical spine
injuries. Anesthesiologists should always have a heightened sense of concern with an
unrelaxed patient fixed in head pins (awake or under general anesthesia).

Indications
The benefits of having a responsive patient during surgery performed near eloquent
cortex are legion. Most commonly, AC is performed on patients with lesions located
near, adjacent to, or even within eloquent cortex. Intraoperative cortical mapping
can be used to identify areas that are dysfunctional as well as those cortical areas
that control important functions such as speech and motor movements. When the
techniques of cortical mapping are combined with an awake, responsive patient,
optimal outcomes can be realized. A prospective comparison in 201113 of AC versus
craniotomies performed under general anesthesia for resection of supratentorial
lesions found a statistically significant improvement in resection quality and better
neurologic outcome in the awake patients.
Certainly a useful technique in adults, AC in pediatric neurosurgical patients
presents the anesthesiologist with a tremendous challenge, because of the differ-
ences in level of cognitive development of children. The youngest patient reported
in the literature to have successfully undergone an AC is a 9-year-old boy, who under-
went resection of a glioblastoma in the left frontotemporal region using propofol seda-
tion.14 Despite this reported success, it is unlikely that many children younger than
10 years are emotionally mature enough to tolerate such a procedure.

Contraindications
Proper patient selection is probably the single most important factor in achieving an
optimal outcome for an AC. Relative contraindications include patient’s age (<11 or
12 years old) and the patient’s general level of emotional maturity. There may be
12-year-old patients who are candidates for an AC, whereas a particular 16-year-
old may not be amenable to such an approach. It is crucial to have thorough discus-
sions with both the patient and their parents before embarking on such an endeavor.
Further, similar conversations should occur with the neurosurgeon and neurologists
involved in order to determine the plan of cortical mapping and what the patient is
required to do.

Patient preparation
As noted earlier, proper patient selection and preparation are paramount. During pre-
operative conversations with the parents and child, the keen pediatric anesthesiolo-
gist is able to get a good assessment of the child’s emotional maturity and
consequent ability to tolerate the proposed procedure. The patient must be able to
calmly and coherently express their needs and concerns to the anesthesiologist,
and there needs to be a certain degree of trust.
86 McClain & Landrigan-Ossar

Technique best practice

A variety of anesthetic approaches are currently used. There is no single best
approach. Each has its advantages and disadvantages. The individual anesthesiolo-
gist needs to have a clear understanding of what the surgical and neurophysiologic
needs are. The specific anesthetic technique may be adopted to best facilitate the
goals within the resource constraints (both human and physical) of the given
environment.
Essentially, the differences in technique boil down to the degree of anesthesia/
sedation used before and after cortical mapping. One approach is to perform the
entire procedure with the patient completely awake.15 This approach necessitates a
mature, motivated, and cooperative child, because several tasks need to be accom-
plished with an awake or minimally sedated patient, including line placement, infiltra-
tion of local anesthetic, placement of head pins, positioning, skull and dural opening,
and resection. This type of approach is well described in the adult literature. A variation
on this technique uses short-acting sedatives and analgesics, such as propofol and
fentanyl, titrated to induce unconsciousness but maintain spontaneous ventilation
for instillation of local anesthetics, insertion of monitoring catheters, placement of
head pins, and skull opening.9,16 Subsequently, patients can be allowed to awaken
to facilitate cortical mapping before definitive resection. After that, the patient may
then have sedatives and opioids reinstituted for the closure.
Another approach is the asleep-awake-asleep technique.17 This approach consists
of inducing general anesthesia and maintaining airway control with a supraglottic
device. General anesthesia is maintained for line placement, placement of head
pins, positioning, and skull and dural opening. The patient is then awakened, the
supraglottic airway is removed, and the surgeons then proceed with mapping and
resection. At the conclusion of the resection, general anesthesia is once again induced
and the supraglottic airway reinserted for closure of the dura, skull, and skin. Although
this technique offers the patient a stress-free procedure up to the point of functional
testing and cortical mapping, there are some disadvantages. Airway management
during emergence and induction while the patient is in head pins presents a greater
challenge. Should the patient cough or buck while immobilized, cervical spine injuries
or scalp lacerations can occur. In addition, brain swelling is a real concern in a patient
who is breathing spontaneously under general anesthesia with an inhalational anes-
thetic and possibly nitrous oxide.

Procedural steps
Line placement and placement of head pins in a child is the first challenge for the anes-
thesiologist. As noted earlier, short-acting, potent sedatives and analgesics such as
propofol and remifentanil may be useful to ensure patient cooperation with placement
of multiple intravenous (IV) lines, arterial lines, and a Foley catheter. These same
agents may also be useful for placement of head pins. Another useful adjunct for scalp
analgesia to facilitate placement of head pins is a scalp block.18 Dosing adjustments
of local anesthetic should be made to account for differing sizes of pediatric patients.
Positioning is best undertaken with an awake patient if at all possible. The advan-
tage to having an awake patient at this juncture is that the patient can communicate
what is most comfortable to them and give feedback as to how to optimize the posi-
tion. A comfortable position is crucial, because the patient needs to remain there for
several hours. It is common for patients to become anxious. Discomfort from awkward
positions only exacerbates this problem. An important aspect of optimal patient posi-
tioning is the creation of a tent or drape tunnel so that the patient does not feel covered
up and can see the anesthesiologist (Fig. 1).
 Challenges in Pediatric Neuroanesthesia 87

Fig. 1. Pediatric patient during an AC. Note creation of tunnel or tent so that patient can
see anesthesiologist and communicate. Note also the head pins in place and the awkward
position that the patient is required to maintain for several hours.

Regardless of the chosen technique, it is most important that the anesthesiologist

themselves (or a trusted designee) remain in close visual contact and verbal commu-
nication with the patient at all times. Communication of every aspect of patient care in
the OR suite must be clear and age appropriate for the patient. It is the anesthesiolo-
gist’s job to decrease the patient’s stress. In this setting, that responsibility involves
letting the patient know who is in the room, what kinds of questions are being asked,
explaining what they may feel, explaining noises, getting feedback about level of
comfort, and distracting the patient to minimize anxiety. This is one of the most chal-
lenging aspects of an AC. It is useful to aggressively control traffic in the room, so that
there is not a constant parade of people of various specialties coming in and out and
inundating the room with multiple conversations. These things can be confusing and
anxiety provoking to a patient who cannot see where the voices and noises are coming
from and may feel like no one is talking to them.
If the patient is sedated during cranial and dural opening and initial exposure, it is
important for them to be awake enough to engage in neurologic testing by the team.
This testing may involve language skills, identifying pictures, memory tests, and so
forth. This is the crucial moment for creating the cortical map to help facilitate resection
of the lesion. If this part of the procedure cannot be accomplished effectively, the whole
point of performing the procedure on an awake patient is rendered moot.
After mapping and resection, the remaining surgical steps involve closure and
removing the head fixation device. This procedure can be performed using sedation
as needed. It is often a welcome break for the patient to be sedated at this point,
because they are often uncomfortable from awkward positions by this point of the
procedure.

Postprocedure care
Caring for patients who have undergone ACs in the postoperative period should have
little difference from postoperative care for pediatric patients who have had craniot-
omies performed under general anesthesia. Frequent neurologic assessments are
the most sensitive indicator of postoperative problems. Control of pain and hemody-
namic parameters is an important goal. The common complications after a craniotomy
performed under general anesthesia are similar to those for ACs: bleeding, cerebro-
spinal fluid (CSF) leaks, electrolyte disturbances, and postoperative emesis. If the pro-
cedure is performed to resect seizure foci, caregivers must be keenly aware of the risk
for postoperative seizures and have a plan for treatment.
88 McClain & Landrigan-Ossar

Avoiding complications
Regardless of the technique chosen, it is important for the anesthesiologist to have
an in-depth discussion with the patient with respect to intraoperative needs and ex-
pectations. The preoperative period is the time to decide whether the patient is a
candidate for an AC. There are no randomized controlled trials comparing the safety
or effectiveness of the techniques just described.
Common complications during ACs include seizures, failure of adequate communi-
cation with the patient, and hemodynamic derangements, including tachycardia and
hypertension.12,19,20 These problems have been noted in larger series of adult ACs.
There are no such data for children. Despite this situation, anesthesiologists should
be prepared to immediately address such concerns in pediatric patients.

iMRI
History of iMRI
The development of iMRI represents a significant leap in the continued effort to
improve intraoperative navigation and the quality of resection of intracranial masses.21
iMRI was first used in the mid-1990s for adult intracranial procedures.22 However, its
success was soon exported to pediatric neurosurgical procedures.23 Coupled with a
frameless navigation system, iMRI offers state-of-the-art ability to localize lesions as
well as improve the quality and pseudoreal time assessment of resection during
neurosurgical procedures. iMRI suites are common, with an increasing number of
such suites being located in pediatric hospitals.24,25 Thus, it is becoming increasingly
important for pediatric anesthesiologists to be familiar with the unique considerations
of working in an iMRI suite.
Before the advent of iMRI, frame-based or frameless navigation systems were used
to aid in localization of intracranial lesions in children. Both types of navigation sys-
tems are predicated on layering a series of fixed points (often using either fiducials
or a face recognition system) over previously obtained imaging. Although these
systems were certainly advantageous, they had their limitations.
First, although these systems offer improved ability to localize lesions that may be
difficult to differentiate with the naked eye (such as in deep brain structures or near
eloquent cortex), they offer no opportunity to intraoperatively assess extent of resec-
tion. iMRI offers high-resolution intraoperative images, which can assess the extent of
resection in pseudoreal time. This technique may lead to improved patient outcomes
and decrease the need for unnecessary second operations in some situations.
Further, these types of navigation systems cannot account for the phenomenon of
brain shift, which occurs naturally during intracranial surgery.26 Brain shift describes
the movement of intracranial structures throughout the procedure. Brain shift leads
to decreased accuracy of navigation because of position changes, egress of CSF
from the cranium, and mass resection. The degree of brain shift can vary with resected
tissue type, patient position, CSF loss, size of craniotomy, hyperventilation, and
amount of tissue resected. As the duration of the surgery increases, so does the degree
of brain shift. This situation leads to the decreased accuracy of conventional intraoper-
ative navigation systems. The advantage that iMRI offers is that images obtained intra-
operatively can update the navigation systems and allow for continued precision.
Equipment
Much like any anesthetic performed to facilitate diagnostic MRI, anesthetics delivered
during procedures that use iMRI require a significant investment in special equip-
ment.27,28 Special MRI-conditional physiologic monitors are required, as are MRI-
conditional anesthesia machines and MRI-conditional infusion pumps. Depending on
 Challenges in Pediatric Neuroanesthesia 89

the particular configuration of a given iMRI suite, it may be possible to use normal sur-
gical instruments, microscopes, drills, tables, and so forth (Fig. 2). However, in suites in
which there is either a movable patient and stationary magnet or a movable magnet
with a stationary patient, the patient still needs to have safe delivery of an anesthetic
during imaging sequences. This procedure necessitates the MRI-safe anesthesia
equipment during the maintenance phase of the anesthetic.
There are a limited number of manufacturers who produce such equipment. The
equipment itself tends to be more expensive and more delicate than standard physi-
ologic monitors, anesthesia machines, and infusion pumps. Also, there are certain
limitations to such equipment. Pulse oximeters are often more sensitive to motion arti-
fact and it can be difficult to obtain a consistent signal in a small child or cold digit.
Electrocardiographic (ECG) interpretation can be challenging, especially during
imaging, when the fluctuating magnetic fields can cause significant interference. ST
analysis can be profoundly unreliable. Core temperature monitoring is not nearly as
simple in an MRI environment. Some manufacturers have produced a temperature
probe, which can be covered with a disposable condom.

Fig. 2. A common iMRI setup. This system uses a movable magnet and stationary patient.
This is the system used at our institution.
90 McClain & Landrigan-Ossar

There are also several pieces of essential equipment that do not have an MRI-safe
or conditional analogue. This equipment includes forced-air warming devices, nerve
stimulators, precordial Doppler probes, and fluid warmers. The lack of availability of
certain MRI-compatible equipment creates challenges in the iMRI environment. This
situation results in the anesthesiologist being responsible for accounting for several
items that must be removed from within the 5-Gauss line when magnet deployment
and imaging occur. The existence of a movable magnetic field creates unique chal-
lenges to the care team. The modern iMRI environment is not analogous to diagnostic
MRI, in which is a stationary field. Because of the movable field, iMRI is unlike any
other environment within which the anesthesiologist works.

Indications
The indications for a pediatric neurosurgical procedure using iMRI can be diverse,
depending on the capabilities of the magnet itself.29 Most commonly, iMRI is used
to help facilitate intracranial tumor resection. iMRI can be used to help localize small
masses, improve precision of other intraoperative navigation systems, and assess
extent of intraoperative resection in pseudoreal time. In addition, iMRI may be used
to help delineate the extent of cortical disconnection in major seizure surgery, such
as corpus callosotomy or functional hemispherectomy.
As imaging technology improves, the opportunities for iMRI to aid in pediatric neuro-
surgical procedures will expand. For example, there are new approaches to ablation of
deep brain lesions using stereotactic MRI-guided laser-induced thermal ablation.30,31
This technique requires multiple intraoperative images, with the ablation completed in
real time while imaging.

Contraindications
A large volume of literature champions the safety of iMRI for many different tumor
types and surgical approaches. However, there is little guidance for practitioners on
exclusion criteria for patients being operated on in an iMRI suite. It may be easiest
to break down the absolute and relative contraindications into some broad categories.
First, we can look at the patient characteristics that may prevent that child from even
entering the iMRI suite. Children who have certain implantable devices may be abso-
lutely excluded from a high-strength magnetic field environment. Examples of these
devices may include implanted cardiac devices (ICDs) (pacemakers and defibrilla-
tors).32,33 Although this remains generally the case, the prevalence of magnetic reso-
nance (MR)-compatible ICDs and pacemakers is becoming more common. Other
devices such as vagal nerve stimulators or certain types of programmable ventricular
shunt valves may be acceptable, depending on the field strength of the magnet.34 It is
crucial to always consult with the MR technologist or radiologist when there is a ques-
tion about the safety of implantable devices.
There are also some relative contraindications because of both the environment of
the room and the size of the bore of the magnet. It can be difficult to perform lengthy
surgical procedures on infants and small children in an iMRI suite for several reasons.
First, the room itself must be kept cold to provide an optimal magnetic field for imag-
ing. Thus, patients such as neonates under general anesthesia, who cannot effectively
self-regulate temperature, are at particular risk for hypothermia and its conse-
quences.35 In our suite, we prefer to avoid having such patients in the iMRI room, un-
less there is some compelling reason to use the magnet and it aids in optimizing the
child’s outcome. In addition, the MRI-conditional equipment itself can be more difficult
to use in small children, in our experience. Movement artifact is more common with
pulse oximetry and ECG. Because of these issues with small children, we prefer not
 Challenges in Pediatric Neuroanesthesia 91

to care for children less than 10 kg or 1 year old, unless there is significant potential
benefit that would alter the risk/benefit ratio in favor of accepting the risks.
Some patients may be too large to safely place in the magnet intraoperatively. Mag-
net bores vary in size, and practitioners must be aware of the limitations of their partic-
ular magnet. Once the patient is under anesthesia and under surgical drapes, they are
not able to alert the care team to potential positioning problems within the magnet. In
order to maintain sterility during intraoperative imaging, the patient remains draped
and padded underneath. This situation further increases the footprint of the patient
(Fig. 3). It is useful to have a template for sizing the patient preoperatively to determine
if larger or obese patients are able to fit in the iMRI. Caregivers should take into ac-
count that the bulk from the drapes and padding is not accounted for in this situation.

Patient preparation
Specific patient preparation for procedures in an iMRI suite should be directed at
screening to ensure the appropriateness of the patient for a high-strength magnetic
field environment. Usually, the MR technician is considered the safety official.36 There-
fore, questions and concerns about the appropriateness of given implantable devices
for a given MRI should be directed to the MR technician. All implanted devices, such
as ventriculoperitoneal shunts, baclofen pumps, aneurysm clips, orthopedic hard-
ware, and so forth, must be noted and cleared by the safety officer. In addition to
the patient, all other people entering the room must also be cleared, including all
medical personnel and parents if it is planned that a parent is to be present during
the induction.
Aside from these considerations, the preparation of the patient for a procedure in an
iMRI suite should proceed similar to the preparation of a patient for the same proce-
dure in a conventional OR suite.

Technique best practice

Although the equipment that the anesthesiologist uses caring for the patient in an iMRI
suite may be different from a conventional OR, the basic approach and anesthetic
techniques should not be significantly different from similar procedures performed
in a conventional OR suite. For craniotomies, common techniques such as total IV
anesthesia or inhaled anesthesia and high-dose opioid are acceptable. No single
best anesthetic technique has been identified for craniotomies.37

Fig. 3. Patient being placed into movable 1.5-T iMRI. Note the degree of draping to pad and
protect the patient as well as maintain sterility.
92 McClain & Landrigan-Ossar

Procedural steps
When caring for patients in an iMRI environment, practitioners must develop pro-
cesses that account for the different risk profile resulting from the movable magnetic
field presented to the patient and caregivers. These additional considerations include
accounting for necessary MRI-incompatible equipment used before placing the
patient in the high-strength magnetic field. Examples include IV needles, airway equip-
ment, manometers, flashlights, scissors, and so forth. Inadvertently leaving
MRI-incompatible equipment around the patient and subsequent placement in a
high-strength magnetic field can result in serious injury or even death to the patient
or caregivers if the object is ferromagnetic and becomes a missile. Even if the object
is not ferromagnetic, it may still present danger to the patient by being a potential
source of thermal injury. Some objects can also cause problems by interfering with
image quality by introducing interference.
Other necessary MRI-incompatible devices include forced-air warming devices and
nerve stimulators. These devices are certainly useful and necessary in modern prac-
tice, but no MRI-compatible version is commercially available. They can be used
safely before placement of the patient in the high-strength magnetic field. Thus, the
issue that arises has to do with the safe use of MRI-incompatible equipment in a
movable magnetic field environment.
One solution is to create a series of checklists. We have used such a system in our
practice with great success. When setting up our iMRI suite, we designed several
different checklists to be implemented at critical steps in the surgical procedure to ac-
count for potential hazards and try to minimize adverse events. The first checklist oc-
curs after induction but before draping the patient. This checklist is designed to ensure
that, after induction, line placement, and positioning, there are no unaccounted for
MRI-incompatible materials near the patient that would be hidden by the drapes
when it comes time to image. After the initial checklist and ensuring appropriate
counts of equipment and instruments, the surgical procedure occurs. Before deploy-
ing the movable magnet, another checklist is implemented. At this point, the nurses
ensure accurate counts of all instruments and sutures. The same process occurs
for the anesthesia provider. The anesthesiologist must ensure that the airway equip-
ment, all IV lines and wires, nerve stimulator, forced-air warmer, fluid warmer, and pre-
cordial Doppler are moved outside the 5-Gauss line. Once all equipment is accounted
for, the magnet is deployed and intraoperative imaging occurs. If repeat imaging is
required after further resection, the process is repeated.

Postprocedure care
Postoperative care of children who have undergone neurosurgical procedures using
iMRI is, in most ways, the same as for those undergoing similar procedures in a con-
ventional OR suite. One caveat is that in suites that take advantage of a high field
strength magnet (eg, 1.5 or 3 T), intraoperative images may obviate routine postoper-
ative MRI. This has certainly been the case at our institution, where we have essentially
eliminated the need for routine postoperative MRI evaluation of children who have un-
dergone craniotomy with iMRI. There remain instances of requests for postoperative
MRI, but these are universally because of some specific concern that would not be
evident on intraoperative images obtained at the end of the case. iMRI has been
shown to help in early diagnosis of some rare serious complications, such as intracra-
nial bleeding, before significant patient compromise.38 Avoidance of such routine
postoperative imaging is advantageous, especially in young children. In addition to
avoiding the logistic challenges of scheduling an MRI in busy centers, young children
may avoid a second anesthetic and the attendant risks.
 Challenges in Pediatric Neuroanesthesia 93

Avoiding complications
Proper training, patient selection, and establishment of strong lines of communication
across disciplines optimize the environment to offer the safest, best possible outcome
to the patient. We began the development of our suite by using a multidisciplinary
approach involving all clinical departments involved in caring for patients in this envi-
ronment: anesthesiology, neurosurgery, radiology, and nursing. We have continued to
have regular meetings of a core group to ensure that this cross-disciplinary commu-
nication is maintained as this technology evolves. This strategy allows us a great
deal of flexibility to address concerns as they arise. These concerns can include prob-
lems with or updates to the iMRI system itself, issues with the anesthesia equipment,
advances in technology that affect the room, near misses, and the use of simulation
approaches to address rare but serious concerns related to the suite.
The use of iMRI is becoming increasingly common. Practitioners must be aware of
the unique considerations when caring for patients in this environment. Its proponents
tout the purported benefits of iMRI. Practitioners must recognize that there are also
some potential downsides as well. Delivering an anesthetic for a patient in diagnostic
MRI is challenging enough. The additional concerns of performing a surgical proce-
dure in a suite with a movable high-strength magnetic field demand a thoughtful
approach from the anesthesiologist.

NEUROINTERVENTIONAL PROCEDURES

It is becoming increasingly common for children with a variety of intracranial diseases

to require anesthesia for diagnostic imaging. In particular, children with some neuro-
vascular lesions require either isolated diagnostic imaging or therapeutic interventions
in the neurointerventional suite. Angiography and embolization are frequent proce-
dures performed in the neurointerventional suite to address a variety of pediatric neu-
rovascular lesions. This section describes the complex considerations of caring for
children with significant neurovascular disease out of the OR.

DIAGNOSTIC CEREBRAL ANGIOGRAPHY

Equipment
Specialized anesthesia equipment is not generally necessary in the interventional
radiology (IR) suite, unlike in MRI. Nonetheless, there are several considerations
worthy of comment. It is generally accepted that older anesthesia equipment is rele-
gated to the non-OR anesthesia (NORA) milieu.39,40 Closed-claims data have shown
that patients in the NORA environment are more likely to suffer more serious harm,
and inadequate monitoring is cited as a contributor.41 Anesthesia equipment in IR,
in which lengthy complex cases are performed on sicker patients, should be standard-
ized with that being used in the main OR. This strategy promotes safety not only by
allowing the use of up-to-date technology for complex patients but also by reducing
the chance of operator unfamiliarity with little-used devices.
Radiation safety in IR is another area of particular concern for anesthesiologists. As
IR anesthesia services increase, anesthesiologists will have increased exposure to ra-
diation. In the procedure room, anesthesiologists’ exposure to radiation is 3-fold
greater than that of radiologists on the opposite side of the procedure table, because
of scatter radiation.42 To reduce exposure, anesthesiologists should wear lead
aprons, preferably wrap-around, and protective eyewear. Portable lead shields should
be placed between the radiation source and the anesthesiologist. If feasible, the anes-
thesiologist should leave the room during angiography runs.43 Portable dosimeters
should be worn and monitored in compliance with local regulations. Because cerebral
94 McClain & Landrigan-Ossar

angiography is generally performed with biplane imaging, the dose to the anesthesi-
ologist is increased, making these precautions even more necessary (Fig. 4).

Indications
Diagnostic cerebral angiography in pediatric patients is generally reserved for those
cases in which computed tomographic angiography or MR angiography provides
partial or questionable information and is the reference standard for diagnosis of neu-
rovascular pathology.44,45 This finding is borne out across several institutions, where
the most common indications are stroke (including moyamoya disease), hemorrhage,
and postoperative evaluation of cerebrovascular treatment.46,47

Patient Preparation
Because this procedure is often brief, little beyond routine preparation for general
anesthesia is required. Preoperative blood pressures are helpful in establishing a
baseline for comparison with intraoperative values. Parents may continue clear oral
fluids until 2 hours before the procedure in the hope that euvolemia contributes to
hemodynamic stability on induction of anesthesia. IV access should be sufficient for
adequate hydration, because the likelihood of significant blood loss is low.

Technique Best Practices and Procedure Steps

Once arterial access is accomplished (usually via a femoral artery), a catheter is
advanced over a guide wire with fluoroscopic guidance to the carotid or vertebral
arteries. Angiography is performed with apnea for the 10-second to 15-second angio-
graphic runs. Once imaging is complete, the arterial sheath is removed, and pressure
is held until hemostasis is achieved (15–20 minutes). Patients are kept flat in the recov-
ery area for several hours to assist with hemostasis.47
General endotracheal anesthesia is most commonly used in pediatric patients.48 A
more mature patient can successfully accomplish the procedure with anxiolysis. How-
ever, sedation beyond anxiolysis can be problematic, because the patient may
become disinhibited and not be able to comply with breath holding.
Fluid management during cerebral angiography is aimed at preserving normoten-
sion by restoring euvolemia after fasting. Euvolemia to slight hypervolemia is prefer-
able to reduce the chances of kidney damage in the setting of a contrast-induced
osmotic diuresis.49,50 Painful stimuli associated with this procedure are minimal

Fig. 4. Room setup for biplane imaging in cerebral angiography. Note radiation sources
both below the procedure table and immediately adjacent to the anesthesiologist.
 Challenges in Pediatric Neuroanesthesia 95

once arterial access is established, and the need for long-acting opioids is low.
Muscle relaxation can be helpful in achieving apnea for angiography but is not
mandatory.

Postprocedure Care
One challenge for pediatric patients is the need for flat bed rest for 2 to 6 hours
after removal of the femoral sheath, to reduce the chances of postprocedural bleeding
or hematoma.51 Behavioral techniques such as reassurance and distraction can be
useful, and chemical adjuncts can help for a younger or less compliant child. Deep
extubation aided by longer-acting narcotics or an a2 agonist, such as clonidine, can
assist a child to sleep quietly in the recovery area.48,52

Avoiding Complications
Complication rates for cerebral angiography in the range of less than 0.4% have been
found in large pediatric series with experienced neuroradiologists.45,46,48 The most
common problem seen is bleeding or hematoma at the femoral puncture, and even
this is rare.53 Neurologic or vascular complications of catheterization are extremely
low in children, as are nephropathic consequences of contrast administration in
patients with no preexisting renal dysfunction.
The risk of acute allergic-like reactions when low-osmolality contrast is used is less
than 1%.54,55 These reactions usually occur within 1 hour of administration and range
from nausea, rash, and mild hemodynamic changes to severe reactions involving ur-
ticaria, bronchospasm, and hemodynamic collapse. Although similar to anaphylaxis,
their nature is still a matter of debate. Mild reactions are treated symptomatically
and are usually self-limited. Treatment of severe reactions is the same as the treatment
of anaphylaxis, with hemodynamic and airway support, antihistamines, steroids, and
epinephrine when indicated.55 In patients at high risk for a reaction (history of previous
reaction, asthma, or atopy), pretreatment with corticosteroids and antihistamines is
standard.

NEUROENDOVASCULAR INTERVENTIONS
Equipment
Equipment requirements and safety considerations for neurointerventional proce-
dures are similar to those required for diagnostic cerebral angiography.

Indications
Indications for neurointerventional procedures include embolization of intracerebral
vascular anomalies, such as arteriovenous malformations, arteriovenous fistulae,
and aneurysms, and targeted injection of intra-arterial chemotherapy for tumors
(Fig. 5).56–59

Patient Preparation
Because of the often lengthy nature of these procedures, meticulous patient posi-
tioning and padding are crucial, because it may be impossible to reposition a patient
once a procedure is under way. Good IV access should be obtained primarily for hy-
dration and drug administration, because the chance of blood transfusion is small. An
arterial line for blood pressure monitoring is usually required for close blood pressure
monitoring as well as regular assessment of activated clotting times (ACTs) if heparin
is administered.
96 McClain & Landrigan-Ossar

Fig. 5. Angiography before (A) and after (B) embolization of right frontal lobe arteriove-
nous malformation in a 9-month-old male.

Technique Best Practices and Procedure Steps

All endovascular work is performed via catheters guided fluoroscopically to the vessel
of interest. Embolic agents such as coils or glue can be deployed or chemotherapeutic
drugs can then be injected. Once the procedure is complete, arterial access is discon-
tinued, and pressure is held at the puncture site. The patient must then lie flat for
several hours to assist hemostasis.
These long procedures are performed almost exclusively under general anesthesia in
pediatric patients. The requirements of an immobile patient and the need for frequent
periods of apnea make this the safest choice. Consistent muscle relaxation is a neces-
sity, because it would be catastrophic for a patient to move during a critical period.60,61
These procedures are not particularly painful, and aggressive pain management may
not be necessary. As in diagnostic angiography, euvolemia to slight hypervolemia
should be maintained to offset the diuretic effect of contrast.49 A continuous infusion
of heparinized saline is used in the guide catheter to prevent microemboli; this can
result in a significant amount of fluid being delivered to the patient via the femoral
sheath, which may not be recognized by the anesthesiologist.62 In addition to this infu-
sion, patients are often heparinized to an ACT of 250 to 300 during the procedure.56
Blood pressure parameters are generally agreed between the anesthesiologist and
neurointerventionalist, and close control of blood pressure is crucial when arteriove-
nous malformations and aneurysms are being treated. Deliberate hypotension has
been described, as has induced asystole to facilitate injection of glue or coil placement
into a high-flow lesion.61 These procedures have not been well described in pediatric
patients, and the technique of balloon-assisted glue embolization may make such
radical measures unnecessary.63
Hemodynamic changes can occur during neurointerventional procedures and
should be monitored closely to maintain stability. Intra-arterial embolization using
Onyx (ev3, Irvine, CA), a popular type of glue, has been reported to induce
bradycardia.64 When a high-flow lesion, which has already affected a patient’s hemo-
dynamics, is closed, it is not uncommon to observe immediate positive changes in the
patient’s condition.56
 Challenges in Pediatric Neuroanesthesia 97

Postprocedure Care
It can be challenging to keep a young child still for several hours after a procedure, and
both chemical and behavioral assistance may be used as described earlier.
After embolization of an intracranial vascular malformation, close control of blood
pressure may be required to reduce the risk of postprocedure hemorrhage.65 This
requirement is particularly true of partial embolization, such as before a surgical resec-
tion. There is evidence for close blood pressure control after the procedure, but there
are few descriptions of how this is to be achieved.66 Our group has had success with a
low-dose dexmedetomidine infusion used after the embolization.

Avoiding Complications
The complications described for diagnostic angiography apply to interventional pro-
cedures as well. Although the most frightening complication for an anesthesiologist
to contemplate is frank rupture of an intracranial vessel, its incidence is estimated
at less than 0.5% of cases.47 Treatment of this complication involves immediate oc-
clusion of the perforated vessel and possible hematoma evacuation, with supportive
treatment as necessary by the anesthesiologist. Migration of embolic agents has been
described; both local flow into unintended vessels and distant glue embolization have
been described. One series described a rate of migration of approximately 3%, with
complications more likely to occur in more complex arteriovenous malformations or
in those with deep draining veins.67 A multispeciality team in the neurointerventional
suite who know how to handle a crisis is an invaluable asset should a complication
arise.68

REFERENCES

1. Walker AA. Neolithic surgery. Archaeology 1997;50(5):19.

2. Lumholtz C, Hrdlicka A. Trephining in Mexico. Am Anthropol 1897;10(12):
389–96.
3. Oakley K, Brooke MA, Akester A, et al. Contributions on trepanning or trepana-
tion in ancient and modern times. Man 1959;59:92–6.
4. Piggott S. A trepanned skull of the beaker period from Dorset and the practice of
trepanning in Prehistoric Europe. Proceedings of Prehistoric Society 1940;6(3):
112–32.
5. Girvin JP. Neurosurgical considerations and general methods for craniotomy un-
der local anesthesia. Int Anesthesiol Clin 1986;24:89–114.
6. Jasper H, Penfield W. Epilepsy and the functional anatomy of the human brain.
1st edition. New York: Little, Brown; 1951.
7. Tasker RR, Marshall BM. Analgesia for surgical procedures performed on
conscious patients. Can Anaesth Soc J 1965;12:29–33.
8. Olsen KS. The asleep-awake technique using propofol-remifentanil anaes-
thesia for awake craniotomy for cerebral tumours. Eur J Anaesthesiol 2008;
25:662–9.
9. Soriano SG, Eldredge EA, Wang FK, et al. The effect of propofol on intraopera-
tive electrocorticography and cortical stimulation during awake craniotomies in
children. Paediatr Anaesth 2000;10:29–34.
10. Ard J, Doyle W, Bekker A. Awake craniotomy with dexmedetomidine in pediatric
patients. J Neurosurg Anesthesiol 2003;15:263–6.
11. Everett LL, Van Rooyen IF, Warner MH, et al. Use of dexmedetomidine in awake
craniotomy in adolescents: report of two cases. Paediatr Anaesth 2006;16:
338–42.
98 McClain & Landrigan-Ossar

12. Nossek E, Matot I, Shahr T, et al. Intraoperative seizures during awake crani-
otomy: incidence and consequences: analysis of 477 patients. Neurosurgery
2013;73(1):135–40.
13. Sacko O, Lauwers-Cances V, Brauge D, et al. Awake craniotomy vs surgery
under general anesthesia for resection of supratentorial lesions. Neurosurgery
2011;68:1192–9.
14. Klimek M, Verbrugge SJ, Roubos S, et al. Awake craniotomy for glioblastoma in
a 9-year-old child. Anaesthesia 2004;59:607–9.
15. Hansen E, Seemann M, Zech N, et al. Awake craniotomies without any sedation:
the awake-awake-awake technique. Acta Neurochir 2013;155:1417–24.
16. Keifer JC, Dentchev D, Little K, et al. A retrospective analysis of a remifentanil/
propofol general anesthetic for craniotomy before awake functional brain map-
ping. Anesth Analg 2005;101:502–8.
17. Hagberg CA, Gollas A, Berry JM. The laryngeal mask airway for awake crani-
otomy in the pediatric patient: report of three cases. J Clin Anesth 2004;16:43–7.
18. Pinosky ML, Fishman RL, Reeves ST, et al. The effect of bupivacaine skull
block on the hemodynamic response to craniotomy. Anesth Analg 1996;83:
1256–61.
19. Skucas AP, Artu AA. Anesthetic complications of awake craniotomies for epi-
lepsy surgery. Anesth Analg 2006;102:882–7.
20. Nossek E, Matot I, Shahar T, et al. Failed awake craniotomy: a retrospective
analysis in 424 patients undergoing craniotomy for brain tumor. J Neurosurg
2013;118:243–9.
21. Manninen PH, Kucharczyk W. A new frontier: magnetic resonance imaging-
operating room. J Neurosurg Anesthesiol 2000;12:141–8.
22. Black PM, Moriarty T, Alexander E, et al. Development and implementation of
intraoperative magnetic resonance imaging and its neurosurgical applications.
Neurosurgery 1997;41:831–42.
23. Nimsky C, Ganslandt O, Gralla J, et al. Intraoperative low-field magnetic reso-
nance imaging in pediatric neurosurgery. Pediatr Neurosurg 2003;38:83–9.
24. Cox RG, Levy R, Hamilton MG, et al. Anesthesia can be safely provided for chil-
dren in a high-field intraoperative magnetic resonance imaging environment.
Paediatr Anaesth 2011;21:454–8.
25. McClain CD, Rockoff MA, Soriano SG. Anesthetic concerns for pediatric pa-
tients in an intraoperative MRI suite. Curr Opin Anaesthesiol 2011;24:480–6.
26. Nimsky C, Ganslandt O, Cerny S, et al. Quantification of, visualization of, and
compensation for brain shift using intraoperative magnetic resonance imaging.
Neurosurgery 2000;47:1070–80.
27. Practice advisory on anesthetic care for magnetic resonance imaging: a report
by the Society of Anesthesiologists Task Force on Anesthetic Care for magnetic
resonance imaging. Anesthesiology 2009;110:459–79.
28. Patteson SK, Chesney JT. Anesthetic management for magnetic resonance im-
aging: problems and solutions. Anesth Analg 1992;74:121–8.
29. Levy R, Cox RG, Hader WJ, et al. Application of intraoperative high-field mag-
netic resonance imaging in pediatric neurosurgery. J Neurosurg Pediatr 2009;
4:467–74.
30. Curry DJ, Gowda A, McNichols RJ, et al. MR-guided stereotactic laser ablation
of epileptogenic foci in children. Epilepsy Behav 2012;24:408–14.
31. Tovar-Spinoza Z, Carter D, Ferrone D, et al. The use of MRI-guided laser-
induced thermal ablation for epilepsy. Childs Nervous System 2013;29(11):
2089–94.
 Challenges in Pediatric Neuroanesthesia 99

32. Pavlicek W, Geisinger M, Castle L, et al. The effects of nuclear magnetic reso-
nance on patients with cardiac pacemakers. Radiology 1983;147:149–53.
33. Erlebacher JA, Cahill PT, Pannizzo F, et al. Effect of magnetic resonance imag-
ing on DDD pacemakers. Am J Cardiol 1986;57:437–40.
34. Crane BT, Gottschalk B, Kraut M, et al. Magnetic resonance imaging at 1.5T
after cochlear implantation. Otol Neurotol 2010;31:1215–20.
35. Bissonnette B. Temperature monitoring in pediatric anesthesia. Int Anesthesiol
Clin 1992;30:63–76.
36. Kanal E, Borgstede JP, Barkovich AJ, et al. American College of Radiology white
paper on MR safety. AJR Am J Roentgenol 2002;178:1335–47.
37. Todd MM, Warner DS, Sokoll MD, et al. A prospective, comparative trial of three
anesthetics for elective supratentorial craniotomy: propofol/fentanyl, isoflurane/
nitrous oxide, and fentanyl/nitrous oxide. Anesthesiology 1993;78:1005–20.
38. McClain CD, Soriano SG, Goumnerova LC, et al. Detection of unanticipated
intracranial hemorrhage during intraoperative magnetic resonance image-
guided neurosurgery. Report of two cases. J Neurosurg 2007;106(Suppl 5):
398–400.
39. Frankel A. Patient safety: anesthesia in remote locations. Anesthesiol Clin 2009;
27:127–39.
40. Melloni C. Anesthesia and sedation outside the operating room: how to prevent
risk and maintain good quality. Curr Opin Anaesthesiol 2007;20:513–9.
41. Metzner J, Posner KL, Domino KB. The risk and safety of anesthesia at remote
locations: the US closed claims analysis. Curr Opin Anaesthesiol 2009;22:
502–8.
42. Anastasian ZH, Strozyk D, Meyers PM, et al. Radiation exposure of the anesthe-
siologist in the neurointerventional suite. Anesthesiology 2011;114:512–20.
43. Dagal A. Radiation safety for anesthesiologists. Curr Opin Anaesthesiol 2011;
24:445–50.
44. Kaufman TJ, Kallmes DF. Diagnostic cerebral angiography: archaic and
complication-prone or here to stay for another 80 years? AJR Am J Roentgenol
2008;190:1435–7.
45. Burger IM, Murphy KJ, Jordan LC, et al. Safety of digital subtraction angiog-
raphy in children: complication rate analysis in 241 consecutive diagnostic
angiograms. Stroke 2006;37:2535–9.
46. Robertson RL, Chavali RV, Robson CD, et al. Neurologic complications of cere-
bral angiography in childhood moyamoya syndrome. Pediatr Radiol 1998;28:
824–9.
47. Alexander MJ, Spetzler RF. Pediatric neurovascular disease: surgical, endovas-
cular and medical management. New York: Thieme; 2006.
48. Wolfe TJ, Hussain SI, Lynch JR, et al. Pediatric cerebral angiography: analysis
of utilization and findings. Pediatr Neurol 2009;40:98–101.
49. Bush WH, Swanson DP. Acute reactions to intravascular contrast media: types,
risk factors, recognition and specific treatments. AJR Am J Roentgenol 1991;
157:1153–61.
50. Mallinckrodt Optiray 240 Package Insert. 2011.
51. Logemann T, Luetmer P, Kaliebe J, et al. Two versus six hours of bed rest
following left-sided cardiac catheterization and a meta-analysis of early ambu-
lation trials. Am J Cardiol 1999;84:486–8.
52. Malviya S, Voepel-Lewis T, Ramamurthi RJ, et al. Clonidine for the prevention of
emergence agitation in young children: efficacy and recovery profile. Paediatr
Anaesth 2006;16:554–9.
100 McClain & Landrigan-Ossar

53. Gross BA, Orbach DB. Addressing challenges in 4 F and 5 F arterial access for
neurointerventional procedures in infants and young children. J Neurointerv
Surg 2013;1–6.
54. Dillman JR, Strouse PJ, Ellis JH, et al. Incidence and severity of acute allergic-
like reactions to IV nonionic iodinated contrast material in children. AJR Am J
Roentgenol 2007;188:1643–7.
55. Pasternak JJ, Williamson EE. Clinical pharmacology, uses, and adverse reac-
tions of iodinated contrast agents: a primer for the non-radiologist. Mayo Clin
Proc 2012;87:390–402.
56. Theix R, Williams A, Smith E, et al. The use of onyx for embolization of central
nervous system arteriovenous lesions in pediatric patients. AJNR Am J Neuro-
radiol 2010;31:112–20.
57. Walcott B, Smith ER, Scott RM, et al. Dural arteriovenous fistulae in pediatric pa-
tients: associated conditions and treatment outcomes. J Neurointerv Surg 2013;
5:6–9.
58. Saraf R, Shrivastava M, Siddhartha W, et al. Intracranial pediatric aneurysms:
endovascular treatment and its outcome. J Neurosurg Pediatr 2012;10:230–40.
59. Shields C, Bianciotto C, Jabbour P, et al. Intra-arterial chemotherapy for retino-
blastoma: report no. 1, control of retinal tumors, subretinal seeds, and vitreous
seeds. Arch Ophthalmol 2011;129(11):1399–406.
60. Schulenburg E, Matta B. Anaesthesia for interventional neuroradiology. Curr
Opin Anaesthesiol 2011;24:426–32.
61. Ogilvy C, Stieg P, Awad I, et al. Recommendations for the management of intra-
cranial arteriovenous malformations. Stroke 2001;32:1458–71.
62. Blanc R, Deschamps F, Orozco-Vasquez J, et al. A 6F guide sheath for endovas-
cular treatment of intracranial aneurysms. Neuroradiology 2007;49:563–6.
63. Andreou A, Ioannidis I, Nasis N. Transarterial balloon-assisted glue embolization
of high-flow arteriovenous fistulas. Neuroradiology 2008;50:267–72.
64. Lv X, Li C, Jiang Z, et al. The incidence of trigeminocardiac reflex in endovas-
cular treatment of dural arteriovenous fistula with onyx. Interv Neuroradiol
2010;16:59–63.
65. Lv X, Wu Z, Li Y, et al. Hemorrhage risk after partial endovascular NBCA and
ONYX embolization for brain arteriovenous malformation. Neurol Res 2012;34:
552–6.
66. Natarajan S, Ghodke B, Britz G, et al. Multimodality treatment of brain arteriove-
nous malformations with microsurgery after embolization with onyx: single-
center experience and technical nuances. Neurosurgery 2008;62:1213–26.
67. Ledezma C, Hoh B, Carter B, et al. Complications of cerebral arteriovenous mal-
formation embolization: multivariate analysis of predictive factors. Neurosurgery
2006;58:602–11.
68. Miguel K, Hirsch J, Sheridan R. Team training: a safer future for neurointerven-
tional practice. J Neurointerv Surg 2011;3:285–7.