The Guidelines for Awake Craniotomy

119

These guidelines are approved by The Japan Neurosurgical Society. The part on anesthetic management is approved bythe Japan Society of Anesthesiology and the part on language assessment during awake craniotomy is approved by theNeuropsychology Association of Japan.Copyright of this English version of the Guidelines is reserved by The Japan Neurosurgical Society.Correspondence to: Takamasa Kayama, MD, Department of Neurosurgery, Yamagata University Faculty of Medicine,2–2–2 Iida-nishi, Yamagata, Yamagata 990–9585, Japan.Phone: ＋81–23–628–5349; FAX: ＋81–23–628–5351.

119

Neurol Med Chir (Tokyo) 52, 119¿141, 2012

The Guidelines for Awake Craniotomy

Guidelines Committee of The Japan Awake Surgery Conference

Preface

Cortical mapping by awake craniotomy has become frequently used worldwide as part of the treatmentstrategy for brain lesions located near language areas. However, no systematic guidelines have been es-tablished for this surgery. The Japan Awake Surgery Conference has now created guidelines for awakecraniotomy for brain lesions near language areas.

The Japan Awake Surgery Conference was established in 2002 for the purpose of continuing researchinto neurocognitive functions as well as establishing and promoting safe methods of awake craniotomy.The 4th annual meeting of this conference decided to establish guidelines for awake craniotomy and or-ganized a guidelines committee. Members specializing in the fields of neurosurgery, neurology, andanesthesiology took part in discussions, a systematic review was carried out, and the guidelines com-mittee attempted to create guidelines in compliance with evidence-based medicine methods as far aspossible. However, the absence of randomized control trials of awake craniotomy forced the guidelinescommittee to use ``de facto standards'' to create the guidelines.

The guidelines consist of three parts: 1) Surgical maneuvers for awake craniotomy, 2) Anestheticmanagement for awake craniotomy, and 3) Language assessment during awake craniotomy. The guide-lines are not intended to override the methods of experienced practitioners, and are not intended to ex-clude methods other than those included. We hope that these guidelines will improve the safety ofawake surgeries and promote the development of the neuroscience of neurocognitive function.

President of The Japan Awake Surgery ConferenceTakamasa KAYAMA, MD

I. SURGICAL MANEUVERS FOR AWAKE CRANIOTOMY

Indications

1. Age[Recommendation]

While there is no specific upper age limit, ananesthesiologist, surgeon, and speech therapistshould consider the condition of each patient care-fully. Surgeons with little experience of awakecraniotomy should try to perform awake surgeryonly in patients aged from 15 to 65 years.

[Commentary]Awake surgery is usually performed in patients

aged from 15 to 65 years. However, patients indicat-ed for such surgery are not only specified by age. Ifthe required tasks can be handled correctly, awake

surgery can be performed in persons younger than15 years and older than 65 years. Patients can un-dergo such surgery at any age if they are consideredto be suitable candidates after other factors havebeen assessed. The cortex is difficult to excite byelectrical stimulation in children aged 7 years oryounger, so they do not fulfill the criteria for corticalelectrical stimulation.1) Patients older than 70 years,who may develop delirium or marked emergent in-crease in blood pressure, require especial attention.

Reference1) Berger MS, Ojemann GA, Lettich E: Neurophysiologi-

cal monitoring during astrocytoma surgery. Neu-rosurg Clin N Am 1: 65–80, 1990

120120

Neurol Med Chir (Tokyo) 52, March, 2012


2. Diseases[Recommendation]

In principle, the indication is for intramedullarydiseases that can be treated surgically.

[Commentary]Epilepsy without macroscopic demarcation be-

tween the normal brain tissue and the lesion, glioma-tosis with indistinct borders, and cavernous heman-giomas that can only be reached through normalbrain regions are typical indications.1) Metastaticbrain tumors are sometimes an indication. Ex-tramedullary tumors such as meningioma are a lesscommon indication, depending on the case.2) For ex-ample, extramedullary tumor corresponding tobrain disease with extended motor nerve involve-ment may be an indication.

References1) Muragaki Y, Maruyama T, Iseki H, Takakura K, Hori

T: [Functional brain mapping and electrophysiologi-cal monitoring during awake craniotomy for intraaxi-al brain lesions]. No Shinkei Geka Journal 17: 38–47,2008 (Japanese)

2) Shinoura N, Yoshida M, Yamada R, Tabei Y, Saito K,Suzuki Y, Takayama Y, Yagi K: Awake surgery withcontinuous motor testing for resection of brain tumorsin the primary motor area. J Clin Neurosci 16: 188–194,2009

3. Sites[Recommendation]

Areas indicated for awake surgery are locationswhere surgical procedures may lead to worsening ofneurological symptoms, but can be assessed by theperformance of intraoperative tasks.

[Commentary]Lesions in and around the anatomical language

areas, lesions in the lateral parietal lobe of thedominant hemisphere (mainly including the angulargyrus), lesions adjacent to the arcuate fibers (su-perior longitudinal fasciculus), lesions adjacent tothe motor cortex, etc.

Awake surgery is indicated for lesions affectingthe triangular and opercular regions of the posteriorpart of the inferior frontal gyrus (Brodmann's areas44 and 45) or the inferior part of the precentral gyruswith respect to the language motor center, as well aslesions in the posterior half of the superior, middle,and inferior temporal gyri of the temporal lobe(areas 41, 42, 22, and 37) or the supramarginal andangular gyri of the parietal lobe (areas 40 and 39)with respect to the sensory language center. Awakesurgery is also indicated for lesions adjacent to thearcuate fibers (superior longitudinal fasciculus) that

appear to connect the motor and sensory languageareas. The hippocampus is located deep inside thetemporal lobe, and is associated with verbal memo-ry, and includes the insular gyri.1) If a lesion is locat-ed near any of the above sites in the dominanthemisphere or if the lesion cannot be confirmed toaffect the nondominant hemisphere, identify thefunctional areas by stimulation.

Reference1) Muragaki Y, Maruyama T, Iseki H, Takakura K, Hori

T: [Functional brain mapping and electrophysiologi-cal monitoring during awake craniotomy for intraaxi-al brain lesions]. No Shinkei Geka Journal 17: 38–47,2008 (Japanese)

4. Other indications such as neurological sym-ptoms[Recommendation and commentary]

The patient has to participate in awake surgery, sothe patient, assessors, surgeons, and anesthe-siologists must all fully understand the meaning ofaggressive resection and possible complications,and be able to recognize whether or not the patientcan tolerate awake anesthesia.

If patients have already developed moderate or se-vere symptoms, mapping and monitoring aredifficult to perform. For example, patients with im-pairment of language functions, such as under-standing, reading, repetition, and object naming, arenot suitable for awake surgery. Among patients whocannot speak fluently but have no disorders of un-derstanding, those with minor naming disordersand decreased word enumeration are candidates,although severe disorders may develop during sur-gery.1)

Patients with serious intracranial hypertensionand those with serious systemic complications arenot suitable.

Reference1) Berger MS, Ojemann GA, Lettich E: Neurophysiologi-

cal monitoring during astrocytoma surgery. Neu-rosurg Clin N Am 1: 65–80, 1990

5. Determination of the dominant hemisphere[Recommendation]

Performance of a provocation test (Wada test) bycerebral angiography is desirable. If determinationof the dominant hemisphere is done by noninvasivetests such as functional magnetic resonance imaging(fMRI), the therapeutic strategy should be defined af-ter considering the possibility of pseudolocalization.

[Commentary]Various advanced procedures such as fMRI, mag-

121121



netoencephalography (MEG), and near infraredspectroscopy (NIRS) have been developed as func-tional tests. These procedures are noninvasive andhave made substantial contributions to neu-roscience and neurology. However, for decision-making about surgical resection, the ``gold stan-dard,'' which is the most reliable procedure availa-ble (the procedure used to define the ``correct an-swer'' as the standard for comparison with newprocedures), should be used. The gold standards foridentification of the dominant hemisphere, func-tional areas, and neuronal functions are the provo-cation test (Wada test), that involves infusion ofanesthetic during cerebral angiography, identifica-tion of functional sites by electrical stimulation, andneurological testing, respectively. Although theanesthetic for the Wada test was amobarbital in theoriginal proposal, propofol is primarily used thesedays because amobarbital is not currently marketedin Japan.3) These ``gold standard'' procedures shouldbe used despite being more invasive because, if lessinvasive but less reliable procedures are used and anincorrect result is obtained, the invasiveness of sur-gery may become greater than necessary. Determi-nation of the dominant hemisphere in patients withtumors causing compression based on fMRI mayhave left-right errors (pseudolocalization) in 14%.4)

The surgical strategy largely depends on whether ornot a lesion affects the dominant hemisphere and in-correct information naturally increases the risk, soperforming the Wada test (the gold standardpreoperative procedure) is considered to be necessa-ry. Although textbooks state that 99% of right-hand-ed persons are left hemisphere dominant, a meta-analysis of 734 patients undergoing the Wada test(including 121 of our patients)2) revealed that thedominant hemisphere for right-handed persons wasthe left hemisphere in 88%, the right in 5%, and bothin 7%.1) The dominant hemisphere determined byelectrical stimulation is the left hemisphere in 91%and the right in 9%. Thus, around 90% of right-hand-ed persons are left hemisphere dominant andaround 10% are right hemisphere dominant, whichis not a low prevalence, suggesting that careful at-tention should be paid during surgery to lesions infunctional areas of the dominant hemisphere. Theresults of the Wada test in left-handed people haveshown that the left dominant:right dominant ratioranges from 1:1 to 3:1, with a slightly higher rate ofleft dominance.

In recent years, the increasing accuracy of nonin-vasive procedures such as fMRI, MEG, and NIRShas provided us with more and more knowledge. Inaddition, when a gradual transition from invasive tononinvasive procedures occurs because of the risk

of complications of cerebral angiography, includingthe Wada test, the risk of pseudolocalization shouldbe accepted. Feedback with respect to comparisonof the results of the Wada test and those of mappingby electrical stimulation is needed.

References1) Ishikawa T, Muragaki Y, Maruyama T, Ono Y, Abe K,

Tanaka M, Amano K, Inoue T, Kubo O, Iseki H, HoriT, Sakai K: [Efficacy of the determination of thedominant hemisphere by Wada test and functionalMRI for tumors around language areas]. Presented atthe 65th Annual Meeting of the Japan NeurosurgicalSociety; October 18–20, 2006; Kyoto (Japanese)

2) Medina LS, Bernal B, Ruiz J: Role of functional MR indetermining language dominance in epilepsy andnonepilepsy populations: a Bayesian analysis. Radiolo-gy 242: 94–100, 2007

3) Takayama M, Miyamato S, Ikeda A, Mikuni N, Taka-hashi JB, Usui K, Satow T, Yamamoto J, MatsuhashiM, Matsumoto R, Nagamine T, Shibasaki H, Hashimo-to N: Intracarotid propofol test for speech and memo-ry dominance in man. Neurology 63: 510–515, 2004

4) Ulmer JL, Hacein-Bey L, Mathews VP, Mueller WM,DeYoe EA, Prost RW, Meyer GA, Krouwer HG,Schmainda KM: Lesion-induced pseudo-dominance atfunctional magnetic resonance imaging: implicationsfor preoperative assessments. Neurosurgery 55: 569–581, 2004

Methods

1. Preoperative treatment1–1. Status and details of simulation[Recommendation]

The tasks that will be performed during surgeryshould be preoperatively rehearsed in the ward.Simulation of the surgical posture, equipment set-up, and role sharing, as well as rehearsal of the tasksfor the patient, surgeons, anesthesiologists, andother surgical staff (such as nurses) should also beperformed in the operating room.

[Commentary]For successful intraoperative mapping with

awake anesthesia, it is important to reduce thepatient's anxiety as much as possible by maintaininga comfortable environment during surgery. Bringthe patient to the operating room on the day beforesurgery, and take enough time to explain what willbe done on the next day (including the posture thatthe patient will be placed in by the surgeon,anesthesiologists, and nurses). Then have the patientactually adopt that posture. If possible, show thepatient a video of surgery on previous patients forbetter understanding. If functional language map-ping is performed, conduct higher function exami-

122122



nation before surgery, perform the tasks that willdone during surgery in the ward in advance, andselect intraoperative tasks, for example, by showingthe patient pictures or photographs of common ob-jects used in object naming and selecting some thatthe patient can answer correctly. If there has been along interval between examination and surgery in apatient with progressive symptoms due to a tumoradjacent to the language areas, the tasks should beselected immediately before surgery.

1–2. Monitoring of anticonvulsants[Recommendation]

In patients who are scheduled to undergo awakesurgery, it is desirable to initiate the administrationof anticonvulsants in advance and maintain effec-tive blood concentrations if enough time is available.Phenytoin can be administered and the concentra-tion increased to the upper limit of the effectiverange (target level 20 mg/dl) by the day before sur-gery.

[Commentary]Even if an effective blood concentration of an an-

ticonvulsant has been maintained since before sur-gery, there is some risk of convulsions during awakesurgery (as described below). Therefore, sufficientpreoperative antiepileptic drug saturation is desira-ble to prevent intraoperative convulsions and foreasy drug loading after the onset of convulsions. Re-garding the selection of drugs, phenytoin is recom-mended, since intravenous administration can beperformed immediately before or during surgerywhen oral administration is not possible, rapid satu-ration is easy, a steady-state blood concentration canbe obtained after a relatively short time (4–5 days),and regulation of the blood concentration is easy.

The bioavailability of phenytoin is high (98%) andthere is little difference between systemic absorp-tion after intravenous and oral administration. Ifthere are 3 or more days before surgery, it is desira-ble to achieve the target blood concentration by oraladministration to reduce patient discomfort. If rapidsaturation immediately before surgery is selected,the target blood concentration can be obtainedpromptly by intravenous administration of pheny-toin.

Especially for patients with tumors located nearthe motor cortex, after obtaining an adequatepreoperative blood concentration of phenytoin, theblood level should be monitored every 2 hours dur-ing surgery. If the concentration is low, intravenousadministration of 250 mg of phenytoin should begiven to raise the concentration to the normal upperlimit (this dose increases the blood level by approxi-

mately 6 mg/dl in a patient weighing 60 kg), or100–200 mg of phenytoin should be given every 4hours during surgery (this dose will increase theblood level by approximately 2.4–4.7 mg/dl in apatient weighing 60 kg).

Sixteen (16%) of the 100 patients who underwentawake surgery at Tokyo Women's Medical Univer-sity from 2004 to the present developed seizures un-der awake conditions, whereas 48 (48%) of these 100patients had a history of seizures before surgery.Twelve (24.5%) of the 49 patients with tumors nearthe motor cortex developed seizures during awakesurgery and this rate was higher than at other sites.Occurrence of seizures during awake surgery is de-fined as clinically obvious convulsions and does notinclude patients who only have afterdischarges.

Among the 80% or more of our 100 patients whohad received preoperative antiepileptic drug ther-apy, patients with lesions near the motor cortex anda preoperative blood level within or above the effec-tive range accounted for 70% of patients both withand without intraoperative convulsions, althoughthe blood level was not measured in all patients.Thus, even if the blood level of an antiepileptic drugis within the effective concentration range, there isno improvement of the preventative effect againstintraoperative seizures, which is more likely to de-pend on the conditions of electrical stimulation.

Preoperative phenytoin loading is not performedat Tokyo Women's Medical University, so its effica-cy has not been demonstrated there. Therefore, thefrequency of intraoperative convulsions in patientswith brain lesions at each site should be comparedwith that determined at institutions where rapidpreoperative phenytoin saturation is performed toassess the usefulness of this procedure. It may alsobe necessary to assess the conditions employed forelectrical stimulation, especially for the motor cor-tex, as well as the use of rapid anticonvulsant satura-tion.

[Saturation]1,2)

Initial loading dose [mg]: target blood concentra-tion [mg/dl] × volume of distribution Vd [l/kg] ×body weight [kg] ＝ 20 × 0.7 × body weight [kg]..(a)

Additional loading dose [mg]: starget blood con-centration － measured value [mg/dl]t× volume ofdistribution Vd [l/kg] × body weight [kg] ＝ (20 －measured value) × 0.7 × body weight [kg] ...........(b)

Where target blood concentration is 20 mg/dl, andVd is specific value for each drug: phenytoin 0.6–0.8(approximately 0.7) l/kg.

For example, in a patient weighing 60 kg, the ini-tial loading dose calculated using (a) is 840 mg,which is administered as three divided doses every 2

123123



to 4 hours. This will avoid cardiovascular adversereactions such as hypotension, bradycardia, and ar-rhythmia, as well as gastrointestinal symptoms.After 12 to 24 hours of administration at the abovedose, initiate therapy at the usual maintenance dose(200–400 mg/day). Measure the blood concentration2 to 3 days after finishing the initial loading dose fororal administration, and at 24 hours after or immedi-ately before surgery when rapid saturation has beenachieved by intravenous administration, and calcu-late the additional loading dose using (b).

The saturation period for phenytoin should be wi-thin the range of 3 to 5 days to avoid the develop-ment of adverse reactions when the blood level ismaintained at the upper limit of the effective con-centration range for too long (1 week or more) beforesurgery. There have been many reports aboutphenytoin-induced skin disorders such as dissemi-nated erythematous papules after approximately 2weeks and serious drug-induced hypersensitivitysyndrome after 2 to 6 weeks in most cases.3)

Because the protein-binding rate of phenytoin ishigh (90–95%), even if the blood level of the drug iswithin the effective range in patients who have a lowserum albumin concentration, the concentration offree drug (not bound to albumin) will be increasedand measured values may not reflect the actual lev-els. If the serum albumin is 3 g/dl or lower, freephenytoin should be simultaneously measured orthe effective phenytoin concentration estimated us-ing the following formula for correction: Measuredphenytoin level/s(0.2 × albumin level) ＋ 0.1t......(c)

Where target range for free phenytoin is 2 to 2.5mg/dl.

References1) Lacy CF, Armstrong LL, Goldman MP, Lance LL:

Lexi-Comp's Drug Information Handbook, 13th ed.Ohio, Lexi-Comp Inc, 2005, pp 1195–1199

2) Ogata H, Masuhara K, Matsumoto Y: [ClinicalPharmacokinetics]. Osaka, Maruzen, 2001, 181 pp(Japanese)

3) Shiohara T, Miyachi Y, Takigawa M (eds): [Dermatolo-gy Practice 19; Master Drug Eruption]. Tokyo, Bunko-do, 2006, 301 pp (Japanese)

2. Various intraoperative methods2–1. Sites and methods of local scalp anesthesia[Recommendation]

For local anesthesia of the scalp, it is common touse long-acting local anesthetics in combinationwith invasive anesthesia and nerve blocks.

[Commentary]Analgesia by local anesthesia is often performed

by the combination of infiltration with local

anesthetic and nerve blocks. At some institutions,anesthesia is performed only by local injection oronly by nerve block. Long-acting local anestheticssuch as ropivacaine and bupivacaine are often usedand these are combined with lidocaine at some insti-tutions.3) Supraorbital nerve block is used if the skinincision is primarily located in the frontal region,whereas auriculotemporal nerve block is used for anincision in the temporal region. Greater or lesser oc-cipital nerve blocks can be added to these blocks. Ifhead fixation is used, an anesthetic is administeredat the sites of the pins in addition to the skin incisionsites. Sufficient anesthetic should be provided at thepin sites because many patients complain of pain atthese sites during emergence.

Preoperative simulation of temporary pseudo-emergence can be performed after fixing the head ina specific posture before the initiation of surgery, toconfirm whether tasks can be performed or whetherthere are any problems with the removal and rein-sertion of a laryngeal mask.1,2)

References1) Fukaya C, Katayama Y: [Intraoperative wake up proce-

dure using propofol total intravenous anesthesia andlaryngeal mask]. No Shinkei Geka Journal 8: 332–337,1999 (Japanese)

2) Fukaya C, Katayama Y, Yoshino A, Kobayashi K,Kasai M, Yamamoto T: Intraoperative wake-up proce-dure with propofol and laryngeal mask for optimal ex-cision of brain tumour in eloquent areas. J Clin Neu-rosci 8: 253–255, 2001

3) Sato K, Kawamata M, Nagata O, Kawaguchi M,Morimoto Y, Kato M, Sakabe T: [Present state ofanesthetic management for awake craniotomy inJapan]. Masui 57: 492–496, 2008 (Japanese)

2–2. Head fixation and posture setting[Recommendation]

Successful awake functional brain mapping/monitoring depends on whether the patient'scooperation can be maintained for a long time.Therefore, head fixation and posture setting are im-portant to keep the patient in a comfortable positionfor a long period.

Although there is no definitive method of head fix-ation and posture setting, continuous feedback is es-sential about whether functional brain mapping/monitoring is successful or not, and whether or notthe patient can comfortably cooperate with surgeryand functional brain mapping/monitoring, and thesurgeon should continue to assess whether theselected method is correct or not.

The basic procedure is as follows:i) Preoperative explanation: It is important to cre-ate an image of surgery for the patient and family. If

124

Fig. 1 Posture setting in the Department of Neurosur-gery of Tohoku University.

124



this is not done, the patient will not understand whatto do and how to cooperate, and will be uneasy dur-ing the surgery. The preoperative explanationshould include basic issues related to brain function-al differentiation, association of the extent of tumorinvasion with functional areas, neurosurgical proce-dures, and functional brain mapping/monitoringprocedures, and be illustrated with pictures, slides,and videos. Also, bring the patient to the actual oper-ating room before surgery, perform head fixationand posture setting, and allow enough time to per-form surgical simulation that includes meeting withthe surgeons and nurses.ii) Head fixation: Whether complete restriction ofmovement of the head by pin fixation or to allowmovement of the head by not performing fixation isbetter has not been decided. The purpose of surgeryis to safely and reliably resect the tumor, and themethod should be established at each institution thatboth maintains patient comfort and allows surgeryto achieve its purpose.iii) Posture setting: To perform functional map-ping of motor and sensory areas including function-al language mapping, craniotomy must extend tosites that include normal brain tissue as well as thetumor. A posture that allows the performance ofwide frontal-temporal-parietal craniotomy is gener-ally used. For posture setting, given that the bodyweight is supported by various parts of the body, in-dividual differences with regard to a comfortableposture and how painful maintaining the sameposture for a long time can be for patients must befully understood. Setting a posture that is only toler-able for a short time and attempting to maintain itfor a long time leads to pain at unexpected sites.How many times the posture can be changed duringsurgery and the patient's desired temperature (hot orcold) must also be confirmed.

[Commentary on approach without head fixation]Posture setting

What surprised us most when we visited the insti-tution of Berger et al., who are pioneers in the use ofthis method for maximum resection of gliomas, isthat Dr. Berger himself took time to carefully seteach patient's posture with pads that were tailor-made for size, shape, and thickness. Their stock ofprepared pads and linen was much larger than ours.Awake surgery provides us with a good opportunityto realize how inadequate our previous posture set-ting is for general anesthesia. The following posturesetting and head fixation procedures are basicallyaccording to the method of Berger et al.1–3)

Preparation on the day before surgery: Bring thepatient to the actual operating room on the day be-

fore surgery, and take enough time to explain whatwill be done on the next day, including posture set-ting. It is important for the patient to meet the sur-geons, assistants, anesthesiologists, and nurses. Atthat time, detailed explanation of the patient's patho-logical condition, and explanation using videosabout tumor resection in combination with awakefunctional brain mapping/monitoring should beprovided to the patient (permission for the use ofvideos should be obtained because they contain per-sonal information). It is as important for the patientto have an understanding of the surgery as it is forthe surgeons to develop an image of the procedure.

To perform functional mapping of motor and sen-sory areas, including functional language mapping,craniotomy needs to expose normal brain tissue aswell as the tumor. In general, to allow for wide fron-tal-temporal-parietal craniotomy, the head is tilted759to the opposite side. The next section coverswhether head fixation should be performed or not.Place a large pad supporting the whole body fromthe shoulder to the waist to avoid torsion of theshoulders and head. To improve venous return,slightly raise the upper part of the body. Find themost comfortable positions for both the upper andlower extremities, and be careful not to overload anypart of the body. It is desirable to fill little emptyspaces with small pads. During posture setting,maintain an environment similar to that during theactual surgery as far as possible, continue conversa-tion, and take time to check for the presence or ab-sence of pain and to be careful not to have any bodypart unsupported (Fig. 1).

To perform mapping, you need to have a clearspace in front of the patient's eyes and to haveenough space to place a portable computer withinthe vision, which is used for object naming in func-tional language mapping. At our hospital, this spaceis created with L-shaped bars and infusion stands

125

Fig. 2 Conditions used for the first case in the Depart-ment of Neurosurgery of Tohoku University (1996).

Fig. 3 Approach with head fixation.

125



(Fig. 2), and transparent drapes may be used to allowvision.6) Surgery takes a long time and osmotic di-uretics such as mannitol may be used because of theinability to employ hyperventilation to control brainswelling, so continuous urine flow is required.

If intraoperative motor functional mapping isdone under general anesthesia, unlike awakeanesthesia, freer setting of the posture and head po-sition (including use of the prone position) is availa-ble. However, functional brain mapping takes timeto perform without administration of muscle relax-ants, so whether the patient will feel comfortable inan unforced posture should be considered when set-ting the posture.

Posture setting on the day of surgery: Even if theposture has already been confirmed on the day be-fore surgery, take enough time to set the patient'sposture again and ensure that he/she is comfortable.To confirm whether the patient feels comfortable ornot, hold a conversation and do not induce anesthe-sia until the completion of posture setting.Head fixation

There is no consensus at this time about whetherhead fixation should be done or not. If we give firstpriority to the patient's comfort, no fixation wouldseem to be more desirable. However, lack of fixationwill lead to constant movement of the surgical field.To continually respond to unexpected movementsfor a long time when manipulating deep brainregions or blood vessels is very stressful for sur-geons. For surgery combined with awake functionalbrain mapping/monitoring, which is based oncooperation and achieving a balance between thesurgeons and the patient, determine whether headfixation should be used or not after careful consider-ation at each institution. Even without head fixation,continuous navigation is available5) by fixing the

reference points to the skull.4)

[Commentary on approach with head fixation]The advantages of head fixation include a fixed

operating field and complete fixation of the conven-tional navigation system, retractor, electroen-cephalograph, or other instruments, so that the sur-geon can operate as under general anesthesia. Thedisadvantages of this approach include more patientdiscomfort compared with the absence of head fixa-tion, due to pain at the pin fixation sites, anddifficulty of moving the body and changing the headposition. Also, treatment of vomiting or convulsionsand reintubation may be difficult, so sufficient simu-lation is necessary. A patient with a left frontal lobeglioma receives 4-point fixation after insertion of alaryngeal mask (Fig. 3). As with 3-point fixation, it isdifficult to rotate and move the head after applica-tion of 4-point fixation, so simulation of emergencysituations is very important.

References1) Berger MS: Malignant astrocytomas: Surgical aspects.

Semin Oncol 21: 172–185, 19942) Berger MS, Kincaid J, Ojemann GA, Lettich E: Brain

mapping techniques to maximize resection, safety,and seizure control in children with brain tumors.Neurosurgery 25: 786–792,1989

3) Berger MS, Ojemann GA, Lettich E: Neurophysiologi-cal monitoring during astrocytoma surgery. Neu-rosurg Clin N Am 1: 65–80, 1990

4) Fukaya C, Katayama Y, Yoshino A, Kobayashi K,Kasai M, Yamamoto T: Intraoperative wake-up proce-dure with propofol and laryngeal mask for optimal ex-cision of brain tumour in eloquent areas. J Clin Neu-rosci 8: 253–255, 2001

5) Leuthardt EC, Fox D, Ojemann GA, Dacey RG, GrubbRL, Rich KM, Ojemann JG: Frameless stereotaxywithout rigid pin fixation during awake craniotomies.Stereotact Funct Neurosurg 79: 256–261, 2002

6) Walsh AR, Schmidt RH, Marsh HT: Cortical mappingand local anaesthetic resection as an aid to surgery of

126126



low and intermediate grade gliomas. Br J Neurosurg 6:119–124, 1992

2–3. Awake state and surgery: Status of anesthe-sia and status of electrical stimulation duringresection[Recommendation]

Resection is often performed under anesthesia orsedation, and the methods vary among institutions.With subcortical mapping, electrical stimulationneeds to be continued even during resection. Thereis no established method at this time and the value ofthe procedure is uncertain.

[Commentary]Although the use of sedation is common during

resection of the lesion, it is done on a case-by-casebasis or never performed at some institutions be-cause it is still controversial. Propofol is often used,but dexmedetomidine, sevoflurane, nitrous oxide,etc., can also be employed.

For subcortical mapping, electrical stimulationneeds to be continued even during resection.However, there is no established method of subcorti-cal stimulation and the methods employed varyamong institutions at present. For subcortical map-ping of the corticospinal tract, the method of record-ing the electromyogram by using Ojemann-typebipolar electrodes as in cortical mapping seems to beemployed relatively often. With bipolar recording,however, the stimulation range is limited to the im-mediate vicinity of the electrodes and injury hasoften already occurred when a response is detected.If the presence of the corticospinal tract cannot bepredicted at a certain distance, subcortical mappingmay well be useless. A method of recording cor-ticospinal motor-evoked potentials (D-wave) thatdescend through the corticospinal tract from spinalepidural electrodes for subcortical mapping is beingdiscussed. The amplitude of D-waves evoked bymonopolar stimulation may reflect the distance be-tween the stimulation points and the corticospinaltract to some degree, so the method of recording theD-wave amplitude under given stimulus conditionsseems to be promising.

Also, in subcortical mapping of language func-tions, conventional bipolar stimulation limits therange as for mapping the corticospinal tract, sowhether language area-derived speech arrest can becertainly detected is unclear and the reliability ofsubcortical electrical stimulation during resection isdifficult to determine.

References1) Duffau H: Contribution of cortical and subcortical

electrostimulation in brain glioma surgery: methodo-logical and functional considerations. NeurophysiolClin 37: 373–382, 2007

2) Fukaya C, Katayama Y, Kobayashi K, Kasai M, OshimaH, Yamamoto T: Impairment of motor function afterfrontal lobe resection with preservation of the primarymotor cortex. Acta Neurochir Suppl 87: 71–74, 2003

3) Katayama Y, Tsubokawa T, Maejima S, Hirayama T,Yamamoto T: Corticospinal direct response in hu-mans: identification of the motor cortex during in-tracranial surgery under general anaesthesia. J NeurolNeurosurg Psychiatry 51: 50–59, 1988

4) Keles GE, Lundin DA, Lamborn KR, Chang EF,Ojemann G, Berger MS: Intraoperative subcorticalstimulation mapping for hemispherical perirolandicgliomas located within or adjacent to the descendingmotor pathways: evaluation of morbidity and assess-ment of functional outcome in 294 patients. J Neu-rosurg 100: 369–375, 2004

5) Nagaoka T, Fukaya C, Katayama Y: [Neurosurgicalprocedures, and electroencephalogram and evokedpotentials, 5. Subcortical mapping]. Rinsho Noha 48:321–325, 2006 (Japanese)


2–4. Conditions and timing of stimulation2–4–1. Cortical stimulation: type of electrode, in-tensity of stimulation, and duration[Recommendation]

In the setting of brain electrical stimulation forfunctional mapping, because of its effect and tomaintain the safety of the brain tissues, the follow-ing methods of using probe electrodes and subduralelectrodes are recommended:Probe electrodes: Interpolar distance of 5 mm anddiameter of 1 mm for bipolar electrodes or diameterof 1 mm for monopolar electrodes; square wavepulses (0.2 or 0.3 or 1 msec) with alternating polarityand frequency of 50 or 60 Hz, and stimulus durationof up to 4 seconds; and current from 1.5 (or 2) mAwith maximum intensity of 16 mA.Subdural electrodes: Interpolar distance of 5 mm to1 cm and diameter of 3 mm for bipolar electrodes;square wave pulses (0.2 or 0.3 msec) with alternatingpolarity and frequency of 50 or 60 Hz, and stimulusduration of up to 10 seconds; and current from 1 mAwith maximum intensity of 16 mA.

[Commentary]Stimulation conditions vary among the types of

electrode and purposes of functional mapping. Therecommendations cover typical methods for corticallanguage mapping. When stimulating the primarymotor area under awake conditions, use of low fre-

127127



quency stimulation or one to five repetitive stimuli isdesirable to prevent convulsions.

For identification of false-positive responses toperipherally spreading electrical stimulation, after-discharges should be monitored.1,2) Corticalexcitability differs between children and adults andalso varies between individuals, so false-negativeresults can occur.

References1) Ebner A, Luders HO: Subdural electrodes, in Luders

HO, Comair YG (eds): Epilepsy Surgery. Philadelphia,Lippincott Williams & Wilkins, 2000, pp 593–596

2) Sanai N, Mirzadeh Z, Berger MS: Functional outcomeafter language mapping for glioma resection. N Engl JMed 358: 18–27, 2008

2–4–2. Subcortical stimulation: type of electrode,intensity of stimulation, and duration[Recommendation]

The conditions for stimulation are the same asthose for ``cortical stimulation.''

Alternative method for stimulation (subcortical):0.2 msec, 50 Hz, stimulus duration of up to 4 sec-onds, from 1 mA to maximum intensity of 20 mA.

Implanted subdural electrode (deep electrode):Used for hippocampal lesions. The interpolar dis-tance is 1 cm or 5 mm.

[Commentary]Experience shows that responses are often identi-

fied by the same tasks and current intensity as withcortical stimulation. If maximum resection is per-formed while checking the response to subcorticalstimulation, 80% of patients develop transient neu-rological symptoms, but 94% of them recover within3 months.1) Subcortical stimulation also allows iden-tification of the following language-related fibers, bywhich various findings have been obtained2,3): su-perior longitudinal fasciculus, arcuate fasciculus,subcallosal fasciculus, inferior fronto-occipital fas-ciculus, inferior longitudinal fasciculus, uncinatefasciculus, orofacial motor fibers, etc.

References1) Duffau H, Capelle L, Denvil D, Sichez N, Gatignol P,

Taillandier L, Lopes M, Mitchell MC, Roche S, MullerJC, Bitar A, Sichez JP, van Effenterre R: Usefulness ofintraoperative electrical subcortical mapping duringsurgery for low-grade gliomas located within eloquentbrain regions: functional results in a consecutiveseries of 103 patients. J Neurosurg 98: 764–778, 2003

2) Duffau H, Capelle L, Sichez N, Denvil D, Lopes M,Sichez JP, Bitar A, Fohanno D: Intraoperative map-ping of the subcortical language pathways using directstimulations. An anatomo-functional study. Brain 125:199–214, 2002

3) Muragaki Y, Maruyama T, Iseki H, Takakura K, HoriT: [Functional brain mapping and electrophysiologi-cal monitoring during awake craniotomy for intraaxi-al brain lesions]. No Shinkei Geka Journal 17: 38–47,2008 (Japanese)

2–5. Treatment of convulsionsRisk: Convulsion can develop during intraopera-

tive mapping and tumor resection.Measures: Reduce the stimulus intensity. Do not

persist with mapping. For tumors near the motorcortex, raise the concentrations of phenobal andphenytoin to the upper normal limits and check thelevels every 2 hours during surgery. If the levels arelower than the limits, appropriately administer 250mg of intravenous phenytoin and 100 mg of in-tramuscular phenobal to increase the concentra-tions to the upper limits. If convulsions occur, putcold water or cold artificial cerebrospinal fluid (e.g.,Artcereb}; Otsuka Pharmaceutical Factory, Inc.,Tokyo) on the brain surface and wait until the con-vulsions cease. If convulsions occur frequently,switch to general anesthesia and then switch back toawake surgery if possible after adequately raisingthe concentrations of anticonvulsants. (For tumorsnear the motor cortex, surgery can continue whilechecking the motor-evoked potentials [second bestmethod].)

2–6. Necessity and usefulness of confirming af-terdischarges: methods and evaluation[Recommendation]

Confirmation of stimulation-induced convulsionsby evaluating the occurrence of afterdischargeson the electrocorticogram should be a basic proce-dure.2) At the very least, until the number of cases ex-perienced by the institution increases, it is essentialthat the stimulation conditions are standardized,and the method of functional evaluation is estab-lished. The risk of mistakenly identifying motor, sen-sory, and language disorders induced by develop-ment of brain dysfunction at distant sites because ofstimulation-induced afterdischarges should beavoided. To confirm whether electrical stimulationis actually being delivered (i.e., the current is flow-ing), electroencephalography is useful. Position theelectrocorticographic electrodes and record theelectrocorticogram without stimulation (Fig. 4).Place small pieces of paper with numbers, etc. onthe brain surface so that surgeons, staff performingelectrophysiological mapping, and staff performinghigher function examination can mutually confirmthe stimulation sites. Stimulate the brain surface for2 to 3 seconds by applying biphasic rectangularpulses of 50 Hz and 0.3 msec pulse width with

128

Fig. 4 Setting the electrocorticograph and electrocor-ticographic stimulation with bipolar electrodes.

128



bipolar electrodes using an inter-electrode distanceof 5 mm. For stable stimulation, the brain surfaceshould be kept moist by using a nebulizer. Increasethe stimulus current from 4 mA in increments of 1mA and determine the optimal current for achievingstable muscle contraction without afterdischarges inthe cortical electroencephalogram. Under the aboveconditions (which we use), effective stimulation isoften obtained at 8–11 mA. Under awake anesthesia,a lower stimulus current is optimal and 4–8 mA isoften used. For the sensory areas, because instan-taneous brain-surface stimulation with a low currentleads to complaints of numbness and discomfort bythe patient, it is desirable to initiate mapping of thesensory areas first. Stimulation is done at the sitespredicted using the neuronavigation system or fromanatomical landmarks such as sulci, gyri, and super-ficial veins, and from the somatosensory-evokedpotentials obtained by median nerve stimulation orlabial stimulation, and the aim is to achieve effectiveresults from the first stimulus. The motor cortexdoes not cover all of the precentral gyrus, but lies be-tween the anteroposterior regions on the side of thecentral sulcus. Therefore, stimulation should be ap-plied along the central sulcus first. Bipolar elec-trodes are used to apply stimulation perpendicular

to the central sulcus. If the craniotomy does not ex-tend as far as the finger area during tumor resectionin the frontal opercular region, you can place stripelectrodes across the central sulcus underneath thedura for stimulation. Electromyography3) is not per-formed for all muscles, so the extremities and facemust be carefully observed at sites where stimula-tion is expected to induce movement.

After completion of mapping of the motor andsensory areas, initiate functional language mapping.First, ask the patient to count from 1 to 50 continu-ously. At this time, increase the stimulus current inincrements of 1 mA, while confirming that there areno afterdischarges in the cortical electroencephalo-gram. A current of up to about 16 mA may be used.Record the sites associated with speech arrest andhesitation. When stimulating the lower portion ofthe precentral gyrus, a negative motor response1)

may inhibit speech. One of the methods for confirm-ing this is to apply stimulation to the brain surface i)while instructing the patient to project the tongueand move it from side to side, ii) while continuingcountermovement of the thumb and forefinger, andiii) while bending and extending the ankle joints. Ifarrest of movement of the tongue or countermove-ment of the fingers and movement of the ankle jointsis observed, the inhibition is associated with a nega-tive motor response and not with language dysfunc-tion. By these procedures, the optimal stimulus cur-rent can be determined and the frontal languageareas identified to some extent. Then, perform ob-ject naming while continuing stimulation.

No abnormalities of counting does not always cor-respond with no disorders of object naming. Showthe patient slides for approximately 2–3 secondseach. Assess whether there is speech arrest, hesita-tion, or wrong answers after presenting the stimu-lus. If these occur, you always need to confirmwhether they are induced by actual stimulation oflanguage function areas, fatigue, inability to see theslides, or the development of seizures. Using the sen-tence pattern for naming ``This is ###'' allows us todetermine whether abnormalities are associatedwith arrest of speech itself by stimulation of the ton-gue motor areas or negative motor areas, or are dueto stimulation of language function areas. Becauseidentification of language areas needs repeated con-firmation of the results, patients have to expend alarge amount of energy. Therefore, functional brainmapping/monitoring requires complete cooperationof the patient.

References1) Nagamatsu K, Kumabe T, Suzuki K, Nakasato N, Sato

K, Iizuka O, Kanamori M, Sonoda Y, Tominaga T:

129129



[Clinical features and significance of negative motorresponse in intraoperative language mapping duringawake craniotomy]. No Shinkei Geka 36: 693–700,2008 (Japanese)

2) Ojemann G, Ojemann J, Lettich E, Berger M: Corticallanguage localization in left, dominant hemisphere.An electrical stimulation mapping investigation in 117patients. J Neurosurg 108: 411–421, 2008

3) Yingling CD, Ojemann S, Dodson B, Harrington MJ,Berger MS: Identification of motor pathways duringtumor surgery facilitated by multichannel electromyo-graphic recording. J Neurosurg 91: 922–927, 1999

2–7. Complications other than convulsions andcountermeasures2–7–1. Pain

Risk: Pain can develop in the skin, muscles, duramater, and sites on the underside of the body.

Measures: Ask the patient about painful sites andtreat with local anesthesia as far as possible. Fix thehead with 3 or 4 pins (if you use a head frame) andplace something soft under the body to allow formovement. For surgery on the temporal lobe, turnthe waist up as far as possible to prevent pain causedby compression of the underside of the waist, whichoften occurs in the lateral position.

2–7–2. Air embolismRisk: Tumors of the inner motor cortex are associ-

ated with a risk of air embolism because the surgicalfield is placed in the highest position.

Measures: Bend the head forward without affect-ing respiration, raise the lower extremities, andbend the abdomen slightly forward to increase thejugular venous pressure. Cover the skull with fibrin,thrombin, and Calcicol immediately after openingthe skull. Keep the head down until the dura mater isopened and then gradually raise the head while ob-serving SaO2. If there are symptoms such as coughand a decrease of SaO2, immediately put the headdown and hold the neck.

2–7–3. Delirium and emotional incontinenceRisk: There are some reports of delirium develop-

ing when anesthesia is stopped to obtain the awakestate. Intraoperative anxiety and pain may alsocause emotional incontinence.

Measures: Avoid decreasing the level of con-sciousness by use of local anesthetic as far as possi-ble. Play the patient's favorite music or take meas-ures to avoid anything that makes the patient un-comfortable so that the patient can undergo surgeryeasily. Depending on the patient's condition and theprogress of surgery, decide whether it should be con-tinued under awake conditions, should be continuedwithout awake conditions, or should be discon-

tinued. If continuation of awake conditions is need-ed, deal with the patient's complaints (primarilypain) as far as possible, but sometimes encourage thepatient to tolerate the discomfort.

2–7–4. Increased intracranial pressureRisk: Increased intracranial pressure may develop

in patients with brain tumors, but seldom in thosewith epilepsy. During awake surgery, arterial carbondioxide tension (PaCO2) tends to be higher than un-der general anesthesia, leading to a higher risk of in-creased intracranial pressure.

Measures: If there is evidence of increased in-tracranial pressure on imaging studies, generalanesthesia should be employed. If awake surgery isconsidered to be absolutely necessary, the decisioncan be made after dural incision following standardintubation. If there is no brain swelling induced byincreased intracranial pressure, switch to awakesurgery after extubation. If brain swelling occursduring awake surgery, consider switching to generalanesthesia.

2–7–5. OthersThere are reports about the development of air

embolism and pneumonia, although whether theseare characteristic problems of awake surgery is con-troversial. Air embolism can be caused by raisingthe head excessively (for example, locating the oper-ating field at the highest position in the motor cortex1). As with general anesthesia, bend the head for-ward without affecting the respiration, raise the low-er extremities, and bend the abdomen slightly for-ward to increase the jugular venous pressure. Coverthe skull with bone wax, fibrin, and thrombin imme-diately after opening the bone. Keep the head downuntil the dura mater is opened and then graduallyraise the head while observing arterial oxygen per-cent saturation (SaO2). If there are symptoms such ascough and decrease of SaO2, immediately put thehead down and hold the neck. Also, for preventionof pneumonia, it seems to be important to preventlowering of consciousness and vomiting (refer to thesection on anesthetic management for details aboutdealing with nausea and vomiting).

Reference1) Shinoura N, Yamada R, Kodama T, Suzuki Y, Taka-

hashi M, Yagi K: Association of motor deficits withhead position during awake surgery for resection ofmedial motor area brain tumors. Minim InvasiveNeurosurg 48: 315–321, 2005

130130



2–8. Decision-making based on the results ofstimulation2–8–1. Epilepsy[Recommendation]

In the case of epilepsy, consider whether theresults of functional brain mapping by electricalstimulation are reliable. Epileptic foci often includefunctional brain sites and the extent of resection in-fluences postoperative seizure control. It is desirableto fully assess the extent and overlap of epileptogen-ic foci and functional sites, and then carefully dis-cuss the indications for resection of functional areasin individual cases depending on the pathologicalcondition.

[Commentary]Epilepsy is a functional disease, and the presence

or absence of functional disorders associated withsurgery influences the indications for surgery. Fordecision-making about additional surgical treatmentand the extent of resection in individual cases, it isimportant to fully understand the pathology ofepilepsy. Assessment of the results obtained by func-tional brain mapping with electrical stimulation re-quires attention to the following points. In patientswith epilepsy, cortical excitability at functional sitesis variable and both false-positive and false-negativeresults of electrical stimulation can occur.1) Dis-placement of brain function sites from their anatom-ical positions can also occur. Therefore, functionalbrain mapping by electrical stimulation should beperformed carefully, and it is desirable to undertakesubdural electrode placement with reference to theresults of various noninvasive physiological tests,such as fMRI, positron emission tomography, andMEG, for detailed mapping. The use of brain surfaceelectrodes allows for complementary cortical func-tion testing to assess the development of symptomsand to measure evoked potentials during voluntaryactivity after electrical stimulation. Epileptogenicfoci often overlap with functional sites in the brain.In such a situation, resection of the focus is superiorto multiple subpial transection and more completeresection results in better postoperative control ofepileptic seizures.2) It is reported that 0% to 63% ofpatients develop persistent functional disorders af-ter resection of functional brain areas involved byepileptic foci. However, due to the small number ofcases, it is unclear whether we should resect all thefunctional sites, whether we should consider theresection of sites with a possible compensatory func-tion, and how to decide on the discontinuation ofresection.

References1) Palmini A, Gambardella A, Andermann F, Dubeau F,

da Costa JC, Olivier A, Tampieri D, Gloor P, QuesneyF, Andermann E, Paglioli E, Paglioli-Neto E, Ander-mann LC, Leblanc R, Kim HI: Intrinsic epileptogenici-ty of human dysplastic cortex as suggested by cor-ticography and surgical results. Ann Neurol 37:476–487, 1995

2) Pondal-Sordo M, Diosy D, T áellez-Zenteno JF, GirvinJP, Wiebe S: Epilepsy surgery involving the sensory-motor cortex. Brain 129: 3307–3314, 2006

2–8–2. Brain tumors[Recommendation]

Functional tissues revealed by mapping should bepreserved unless consent giving priority to resectionis obtained or the surgeon determines that resectionis feasible. Accumulate experience with mappingand pay careful attention to false-positive results(nonfunctional brain tissues despite positive find-ings on stimulation).

[Commentary]Intraoperative functional testing during awake

surgery involves mapping by electrical stimulationand monitoring to observe neurological findings.Mapping is performed to identify functional braintissues and to prevent neurological complicationsinduced by resection and damage to functional tis-sues during brain tumor removal. Therefore, thesites where symptoms occur during mapping shouldbe preserved in principle because they are likely tobe functional tissues. If they are not preserved, wesee no point in performing awake surgery. However,if tumors coexist with functional tissues2) andpreoperative consent has been obtained, functionaltissues may not be preserved if the decision is con-firmed to give priority to tumor resection after ac-cepting the risk of complications and the fact thatpostoperative symptoms are likely to develop evenwith a response (e.g., negative motor areas in thesupplementary motor cortex). Because responses aresometimes false-positive, you should acquireproficiency in mapping. False-positive findings areprimarily obtained because awake conditions arepoor and do not allow patients to perform theirtasks, and sometimes because the basic conditionsfor the tasks are poor (e.g., the patient cannot see thescreen because a drape covers his/her eyes). At ourhospital, some patients could not perform the nam-ing task due to inability to see the screen because ofconjugate deviation induced by stimulation of oculo-motor fibers, but they were regarded as having lan-guage arrest (the truth was recognized by reviewingvideos). Recently, Berger et al.1) have insisted on thevalidity of a ``negative mapping strategy,'' suggest-

131131



ing that language areas do not have to be identifiedas a positive control (and major craniotomy does nothave to be performed for identification), and thatresection can be performed if language areas do notexist within the resection zone under certain condi-tions (60 Hz, maximum 6 mA). Given that this is areport from the most experienced institution, resec-tion after identifying the language areas seems to besafer at less experienced institutions. It is significantthat their report indicates that awake language map-ping allows us to perform even aggressive resectionwith a very low incidence of complications and that

a report on language mapping was published in atop clinical journal, suggesting that evaluation of itsusefulness as a surgical procedure for glioma hasbeen established.

References1) Sanai N, Mirzadeh Z, Berger MS: Functional outcome

after language mapping for glioma resection. N Engl JMed 358: 18–27, 2008

2) Skirboll SS, Ojemann GA, Berger MS, Lettich E, WinnHR: Functional cortex and subcortical white matterlocated within gliomas. Neurosurgery 38: 678–685,1996

II. ANESTHEIC MANAGEMENT FOR AWAKE CRANIOTOMY

1. IntroductionIn the 1800s, resection of foci in epileptic patients

was performed by craniotomy under local anesthe-sia.6) With no electroencephalogram, direct stimula-tion of the cortex was employed to detect the epilep-tic focus and identify functionally important sites,which seems to be the prototype of current awakecraniotomy. In the 1900s, with addition of sedation,surgery became more comfortable for patients.8)

Using codeine, thiopental, and meperidine, manage-ment was conducted under spontaneous respirationor partially by tracheal intubation. Epileptic surgerythen came to emphasize intraoperative electroen-cephalography.9) In the 1960s, neuroleptanalgesiawas introduced into anesthesia, and the combina-tion of droperidol and fentanyl was considered espe-cially useful for surgery in patients with temporallobe epilepsy because it had less influence on the in-traoperative electroencephalogram.5) In addition,the development of a long-acting local anesthetic,bupivacaine, facilitated awake craniotomy. As aresult, many procedures for intractable epilepsy em-ployed neuroleptanalgesia.1,7) Thereafter, short-act-ing analgesics such as sufentanil and alfentanil wereintroduced.4) Propofol was introduced for awakecraniotomy because it is short-acting and has an-ticonvulsant and antiemetic effects.10) It is nowwidely used as the main sedative. Recently, newanesthetics such as dexmedetomidine2) and remifen-tanil3) have been introduced, while use of a laryngealmask has been initiated for airway management.11)

Because procedures that are not necessary duringordinary general anesthesia, such as airway manage-ment and treatment of intraoperative convulsions,are required, we would like you to refer to theseguidelines for reliable and safe anesthetic manage-ment. There is limited evidence about anestheticmanagement during awake craniotomy, so themethods in actual use and those recommended bythe review committee are presented. Also, because

there is ``awake surgery'' in the cardiac surgery fieldand the ``wake-up test'' in orthopedic surgery, weare using the term ``awake craniotomy'' here to dis-tinguish awake brain surgery from those proce-dures.

References1) Archer DP, McKenna JMA, Morin L, Morin L,

Ravussin P: Conscious-sedation analgesia duringcraniotomy for intractable epilepsy: a review of 354consecutive cases. Can J Anaesth 35: 338–344, 1988

2) Bekker AY, Kaufman B, Samir H, Doyle W: Use ofdexmedetomidine infusion for awake craniotomy.Anesth Analg 92: 1251–1253, 2001

3) Berkenstadt H, Perel A, Hadani M, Unofrievich I,Ram Z: Monitored anesthesia care using remifentaniland propofol for awake craniotomy. J NeurosurgAnesthesiol 13: 246–249, 2001

4) Gignac E, Manninen PH, Gelb AW: Comparison offentanyl, sufentanil and alfentanil during awakecraniotomy for epilepsy. Can J Anaesth 40: 421–424,1993

5) Gilbert RGB, Brindle GF, Galindo A: Anesthesia forneurosurgery. Int Anesthesiol Clin 4: 842–847, 1966

6) Horsley V: Brain-surgery. Br Med J 2: 670–675, 18867) Manninen P, Contreras J: Anesthetic considerations

for craniotomy in awake patients. Int AnesthesiolClin 24: 157–174, 1986

8) Pasquet A: Combined regional and general anesthe-sia for craniotomy and cortical exploration. II.Anesthetic considerations. Curr Res Anesth Analg 33:156–164, 1954

9) Penfield W: Combined regional and general anesthe-sia for craniotomy and cortical exploration. I. Neu-rosurgical considerations. Curr Res Anesth Analg 33:145–155, 1954

10) Silbergeld DL, Mueller WM, Colley PS, Ojemann GA,Lettich E: Use of propofol for awake craniotomies.Surg Neurol 38: 271–272, 1992

11) Tongier WK, Joshi GP, Landers DF, Mickey B: Use ofthe laryngeal mask airway during awake craniotomyfor tumor resection. J Clin Anesth 12: 592–594, 2000

132132



2. Basic policyi) Communicate with the surgeons and operatingroom staff based on a detailed surgical and anesthet-ic plan.ii) To handle intraoperative respiratory problemsand rapidly changing risks, management and super-vision by anesthesiologists who have acquired ex-tensive experience with awake craniotomy is re-quired.iii) To safely manage rapidly changing intraopera-tive conditions, ensure that backup anesthesiologistsare available in addition to the attending anesthe-siologists.iv) To allow smooth switching to general anesthe-sia if the anesthesiologist considers it difficult tocontinue awake craniotomy, establish a system forcooperation with the surgeons and operating roomstaff.v) Do not use inhalational anesthetics that are ab-sorbed and excreted by the respiratory system be-cause a definitive airway is not established. (Inhala-tional anesthetics that could possibly cause an in-crease of brain volume should not be used becauseof the uncertain management of PaCO2 duringawake craniotomy.) Use propofol as the basic seda-tive.vi) Because management is performed under spon-taneous respiration, carefully titrate the sedativeand analgesic drugs. Maximize the use of localanesthesia for analgesia. (During the unconsciousperiod, management with controlled respiration viathe laryngeal mask airway [LMA], etc. is available.)vii) Take measures to prevent nausea and vomitingthat could lead to respiratory complications.viii) Electrical stimulation during functional map-ping may induce convulsions, occasionally resultingin inability to continue the procedure, which thenrequires rapid countermeasures to be taken.

[Commentary]Awake craniotomy was used for surgical treat-

ment of epilepsy in the early 20th century, and thenwas applied to surgery for brain tumors, cerebral ar-teriovenous malformations, and cerebral aneurysmsassociated with important areas such as the motor orsensory cortex and language cortex.2,3) The purposeof awake craniotomy is to prevent brain dysfunctioninduced by surgery and to precisely resect the dis-ease focus in order to improve the patient's progno-sis and quality of life. The purpose of anestheticmanagement is, while putting the patient's safetyfirst, to remove psychophysical pain and allow thenecessary surgery to be done. Each of the sections inthese guidelines and the corresponding commentarydescribe the details of anesthetic management for

awake craniotomy. Because there have not beenenough randomized controlled studies of anestheticmanagement for awake craniotomy, managementthat is not based on such evidence in these guide-lines is based on the methods recommended by insti-tutions familiar with awake craniotomy. Therefore,if anesthetic management based on randomized con-trolled study evidence is reported in the future, theseguidelines will be appropriately reviewed.

For successful awake craniotomy, the first point isthat the patient's cooperation is essential. Second,preoperative and intraoperative communication andagreement among neurosurgeons, anesthesiologists,and operating room staff familiar with awakecraniotomy is required. For anesthetic management,it is necessary to establish the airway, stabilizehemodynamics, and prevent increase of intracranialpressure. Because management of PaCO2 is moredifficult during awake craniotomy, inhalationalanesthetics that could possibly increase brainvolume should be avoided. Hence, sedation andgeneral anesthesia with propofol is currently thestandard for awake craniotomy. While the patient isunconscious, respiratory management with an LMAcan be used. Each of the sections in these guidelinesdescribes the details of respiratory management.

During awake craniotomy, because scalp blockand infiltrational anesthesia for sufficient pain con-trol require a large volume of local anesthetics, cau-tion should be paid with regard to local anesthetictoxicity.1) During awake craniotomy, it is also neces-sary to prevent adverse reactions such as nausea,vomiting, and convulsions, and to deal with suchreactions immediately if they occur. If establishmentof an airway is difficult or if other adverse reactionsinterfere with the patient's safety, after discussionbetween the anesthesiologists and neurosurgeons,awake craniotomy should be speedily discontinuedand switching to general anesthesia should be consi-dered.4)

References1) Archer DP, McKenna JMA, Morin L, Ravussion P:

Conscious sedation analgesia during craniotomy forintractable epilepsy: a review of 354 consecutivecases. Can J Anaesth 35: 338–344, 1988

2) Burchiel KJ, Clarke H, Ojemann GA, Dacey RG, WinnHR: Use of stimulation mapping and corticography inthe excision of arteriovenous malformations in sen-sorimotor and language-related neocortex. Neurosur-gery 24: 322–327, 1989

3) Duffau H, Capelle L, Sichez JP, Faillot T, AbdennourL, Law Koune JD, Dadoun S, Bitar A, Arthuis F, VanEffenterre R, Fohanno D: Intra-operative direct elec-trical stimulations of the central nervous system: theSalp âetri àere experience with 60 patients. Acta Neu-

133133



rochir (Wien) 141: 1157–1167, 19994) Piccioni F, Fanzio M: Management of anesthesia in

awake craniotomy. Minerva Anesthesiol 74: 393–408,2008

3. Premedicationi) To allow complete intraoperative emergence, donot administer premedication that could possiblycause residual sedation.ii) If there is no choice but to administer premedi-cation, use a benzodiazepine that could possiblyproduce antagonism.iii) Make a decision about premedication with an-ticonvulsants after consulting the patient's physi-cian.

[Commentary]During awake craniotomy, it is important for

patients to be sufficiently awakened during surgeryto perform language and motor tasks that yield relia-ble results, based on which the extent of resection isdetermined. Therefore, as a matter of principle,drugs that could affect emergence should not be ad-ministered. For successful awake craniotomy, it iscrucial to build a relationship of trust among thepatient, the surgeons, the anesthesiologists, and theoperating room staff.1) The establishment of such apatient-centered relationship reduces the need forsedatives. However, if sedatives have to be ad-ministered, benzodiazepines are recommended asantagonists are available. If surgery is being done fora tumor, hypercapnia induced by sedation can possi-bly result in an increase of intracranial pressure,and this requires special caution.

Convulsions are one of the most significant com-plications of awake craniotomy. Difficulty in ven-tilating the patient when convulsions persist andrespiratory arrest occurs can lead to a fatal outcome.Because the patient's condition needs to be consi-dered, preoperative administration of anticonvul-sants should only be done after discussion with theattending physician. Note that propofol also has ananticonvulsant effect. With regard to other drugssuch as H2 blockers that are administered duringgeneral surgery, the policy of the institution shouldbe adopted.

Among antiemetics, metoclopramide hydrochlo-ride (Primperan}; Sanofi-Aventis K.K., Tokyo) is notrecommended because of potential adverse effectscaused by enhanced peristalsis. Some reports of dex-amethasone being administered to control the in-tracranial pressure and prevent vomiting have beendescribed in other countries. However, this proce-dure is not covered by health insurance in Japan.Also, propofol has a useful antiemetic effect.

Reference1) Whittle IR, Midgley S, Georges H, Pringle AM, Taylor

R: Patient perceptions of ``awake'' brain tumor sur-gery. Acta Neurochir (Wien) 147: 275–277, 2005

4. Basic monitoring and preparationi) Monitor the electrocardiogram, invasive arterialblood pressure, percutaneous oxygen saturation, ex-piratory partial pressure of carbon dioxide, urinevolume, and body temperature.ii) Create peripheral venous access for continuousadministration of anesthetics and blood transfusion.iii) Procedures are performed without a secure air-way, so careful respiratory management is needed.Management can be done with spontaneous breath-ing or assisted ventilation via the LMA, etc.

[Commentary]Basically, comply with the guideline of the

Japanese Society of Anesthesiologists for installa-tion of monitors. If sedation is used during awakecraniotomy, regardless of the method of establishingthe airway, different precautions from those duringtracheal intubation should be taken. If a device suchas the LMA is not used to establish the airway, pre-cise management of PaCO2 is difficult and carefulobservation is necessary to assess the frequency ofbreathing and the presence or absence of forcedrespiration. Therefore, for good anesthetic manage-ment, an environment that allows easy observationof the respiratory status is required, including use oftransparent drapes to allow sufficient observation ofthe patient's mouth, neck, and chest. A stethoscopetaped to the chest wall is one method. Some nasalcannulae have ports for measurement of expiratorycarbon dioxide, but sometimes accurate data are notobtained, indicating limited usefulness. For manage-ment with an LMA, even if spontaneous ventilationis maintained, measure expiratory carbon dioxidetension and support respiration if necessary. Venti-lation does not often allow for sufficient respiratorymanagement, so an arterial cannula should beplaced for easy arterial gas analysis when needed.

There have been reports of air embolism duringawake surgery. Especially when surgery is per-formed under spontaneous respiration without anLMA, caution is required. In this case, end-tidal car-bon dioxide is not so useful as described above, andit is difficult to determine whether changes of satu-ration of peripheral oxygen result from worsening ofrespiratory status or air embolism.2,5) Upper airwayobstruction causes lower negative intrathoracicpressure and may result in increased risk of air em-bolism. Although a bispectral index (BIS) monitor isgenerally considered to be useful, its value is limited

134134



and is not strongly recommended in the field of neu-rosurgical anesthesia. There have been reports that aBIS monitor is useful even if placed at sites otherthan the forehead. However, because shaving of thehead recently tends to be minimized, the site wherethe monitor can be placed is actually limited to theforehead. Even if a pediatric BIS sensor is used, dis-infectant or blood often enters the connections ofthe sensors and interferes with analysis. In addition,a BIS monitor is mainly useful on emergence, and itsvalue before that is limited by contamination fromelectromyogram signals and noise from electric scal-pels used during craniotomy.

Maintenance of the airway during awakecraniotomy is done by one of two methods; onemethod is to depend on spontaneous respirationwith no devices and the other method is to use adevice such as the LMA. If a device is used, we canrely on spontaneous respiration and, if necessary,provide respiratory support, or we can actively per-form ventilation. In Japan, the LMA has tended to beused recently. Tracheal intubation is not recom-mended since it is likely to interfere with an awakestudy because of complications caused by emer-gence-induced coughing and depressed laryngealfunction including hoarseness. Although a trachealtube can be placed in the pharynx via the nose forrespiratory support, if necessary, or emergency intu-bation can be done with a bronchoscope, nasalbleeding can become a problem. If management isperformed by spontaneous respiration withoutdevices, hypercapnia will become a problem, butcan often be dealt with by target controlled infusion,avoiding narcotics, or large craniotomy.1,3,4) Partlyto decrease the dose of narcotics, it is important toadminister sufficient local anesthesia as describedin the next section.

If 3-point fixation is performed, it is assumed thatreinserting the LMA or tracheal intubation will bedifficult in some cases. It is important to testremoval and reinsertion of the LMA, to sufficientlydiscuss measures to tracheal intubation when the ag-gravation of the breathing state occurs during sur-gery. Also, when 3-point fixation is employed, checkwith the surgeons that the neck has not been twistedor anteflexed. As it is known that more problemsrelated to the respiratory tract and respiration de-velop in patients with a body mass index À30, becareful about deciding to perform awake cranio-tomy in such patients.6)

References1) Herrick IA, Craen RA, Gelb AW, Miller LA, Kubu CS,

Girvin JP, Parrent AG, Eliasziw M, Kirkby J: Propofolsedation during awake craniotomy for seizures:

patient-controlled administration versus neuroleptanalgesia. Anesth Analg 84: 1285–1291, 1997

2) Morimoto Y, Sakabe T: [Practical side of awakecraniotomy—primarily anesthetic management forstereotactic surgery], in Furuya H (ed): [Practice ofAwake Craniotomy—Main Points of AnestheticManagement]. Tokyo, Shinko Trading Co., Ltd. Publi-cation Department of Medical Books, 2003, pp 38–60(Japanese)

3) Nagaoka Y, Miura Y, Takaoka S, Iwabuchi M, Amaga-sa S, Kotani N: [Considerations for awake craniotomyin our hospital—especially respiratory complicationsand brain edema]. Rinsho Masui 29: 836–840, 2005(Japanese)

4) Sarang A, Dinsmore J: Anaesthesia for awakecraniotomy—evolution of a technique that facilitatesawake neurological testing. Br J Anaesth 90: 161–165,2003

5) Scuplak SM, Smith M, Harkness WF: Air embolismduring awake craniotomy. Anaesthesia 50: 338–340,1995

6) Skucas AP, Artru AA: Anesthetic complications ofawake craniotomies for epilepsy surgery. AnesthAnalg 102: 882–887, 2006

5. Admission, induction, and local anesthesiai) Initiate oxygen delivery after validation of thevital signs with a patient monitor.ii) Induce anesthesia with only propofol, or incombination with fentanyl/remifentanil. A target-controlled infusion (TCI) system should be used forpropofol administration to precisely manage the lev-el of sedation. The administration of fentanyl beforeemergence should be minimal.iii) Maintain general anesthesia under spontane-ous respiration with a facemask, or under as-sisted/controlled ventilation after insertion of anLMA.iv) Insert a urethral catheter.v) Provide effective analgesia with local anesthet-ics by infiltration at the site of skin incision, and/orby selective nerve blocks of, for example, thesupraorbital nerve and the greater occipital nerve.Ropivacaine is commonly used as a long-acting localanesthetic.

[Commentary]Management of general anesthesia with inhala-

tion anesthetics is ineffective when the airway ispoorly established. The main impediments are un-certain delivery of the anesthetics and hazardouscontamination of the operating room with theanesthetics. Propofol should be used as a hypnoticagent. Propofol is an intravenous anesthetic drugthat permits faster and clearer emergence than inha-lation anesthetics, which affect the electroen-cephalogram and sometimes induce excitement at

135135



emergence from anesthesia. A TCI system canreasonably be used for propofol in order to maintainan optimal hypnotic level by adjusting the effect-siteconcentration of the agent, as the sedative effect de-pends on the effect-site concentration. In the case ofanesthetic management without the TCI system, thepropofol administration should preferably bemanaged by continuous infusion combined withrepetitive injection based on the effect-site concen-tration calculated with pharmacokinetic simulation.Opioids are sometimes used for sedation, but givesome residual effects to the consciousness level afteremergence. Remifentanil is therefore suitable in themanagement of strong surgical stimulation beforeemergence, as its effect rapidly disappears. It is alsoreasonable to administer small dose of fentanylrepeatedly, expecting only a slight residual analgesiceffect.

The airway before emergence is usually managedwith a facemask or an LMA. Airway managementunder assisted/controlled ventilation and spontane-ous respiration can be performed safely with LMA,though it is generally difficult to extubate safely andsmoothly at awakening. The incidence of muscleweakness in the conscious state is rare with LMA, asmuscle relaxants are not necessary for LMA inser-tion. A nasogastric tube should not be inserted, as itleads to discomfort in the pharynx, nausea, andvomiting during the conscious state. Remove thenasogastric tube before emergence if it has to be in-serted during general anesthesia. Insert a urethralcatheter after induction of anesthesia, as the opera-tion will take a long time.

The key to anesthetic management for awakecraniotomy is to achieve a ``pain-free'' state withmultimodal pain management. Since intravenousanesthetics affect the state of consciousness andrespiration, local anesthetics are essential for as-sured analgesia. This is achieved with the use oflong-acting local anesthetics such as ropivacaine orbupivacaine, or lidocaine combined with epine-phrine. Problems such as local anesthetic toxicitydid not occur even at a mean ropivacaine dose of 3.6mg/kg in a study of the blood concentration of localanesthetics for awake craniotomy. Local anestheticsare administered by infiltration around pin fixationand the site of the skin incision, along with selectivenerve blocks (supraorbital nerve, greater occipitalnerve, etc.) Gauze soaked with local anesthetics canbe pressed against the wound. Because direct con-tact of local anesthetics with the brain parenchymacauses central nervous system symptoms such asconvulsions, the administration of local anestheticsafter dural incision should be performed carefully.

References1) Costello TG, Cormack JR, Hoy C, Wyss A, Braniff V,

Martin K, Murphy M: Plasma ropivacaine levels fol-lowing scalp block for awake craniotomy. J NeurosurgAnesthesiol 6: 147–150, 2004


6. Before emergencei) In principle, sedative and analgesic drugs shouldnot be used during the awake time. Check the sur-geons' preference for the level of consciousness(level of sedation).ii) Discontinue propofol after the dural incisionhas been made. If sedation is to be continued, pro-vide the required dose of propofol, etc.iii) Closely monitor the patient because body move-ments may occur suddenly during the course ofemergence.iv) If an LMA and a gastric tube are used, confirmspontaneous respiration before removal.v) If the patient exhibits restlessness and cannotkeep still, intraoperative emergence may be aban-doned after discussion with the surgeon and theprocedure may instead be performed under stan-dard general anesthesia.

[Commentary]Because the tasks and tests used for brain func-

tional mapping and electrocorticography to deter-mine the extent of epileptic focus resection aregenerally susceptible to sedative and analgesicdrugs, in principle, such drugs should not be ad-ministered during the awake time. Since even anal-gesics administered before emergence influence theextent of emergence, check the neurosurgeon'spreference about the tests and sufficiently controlthe depth of anesthesia while considering thepatient's preoperative condition. Poor emergencemay make functional assessment difficult.4)

During the period of strong surgical stimulation,including scalp incision, muscle detachment, andremoval of the bone flap and dural incision, provideadequate sedation and analgesia, and discontinuepropofol on completion of dural manipulation.2)

Body movement sometimes occurs during emer-gence as with other surgical anesthesia practice. Be-cause sudden body movement can be more harmfulwhen the skull is fixed with head pins and opened,sufficient vigilance is required and anesthesiologistsshould be prepared to control body movement.Anesthesiologists should promptly control majorchanges in circulatory and respiratory systems,

136136



which often occur during this period.3) When a gas-tric tube or LMA is used, check spontaneous respira-tion and remove on emergence. Due to restlessness,the patient may not remain still and cooperate withfunctional tests.1) If restlessness is considered to becaused by excitation, pain, poor posture, low tem-perature, residual anesthetic, or a painful urethralcatheter, deal with the cause. If the cause isunknown or cannot be controlled, after discussionwith the attending surgeon, intraoperative emer-gence may be abandoned and surgery may be dis-continued or performed under general anesthesia.

References1) Berkenstadt H, Perel A, Hadani M, Unofrievich I,

Ram Z: Monitored anesthesia care using remifentaniland propofol for awake craniotomy. J NeurosurgAnesthesiol 13: 246–249, 2001

2) Huncke K, Van de Wiele B, Fried I, Rubinstein EH:The asleep-awake-asleep anesthetic technique for in-traoperative language mapping. Neurosurgery 42:1312–1316, 1998

3) Keifer JC, Dentchev D, Little K, Warner DS, FriedmanAH, Borel CO: A retrospective analysis of a remifen-tanil/propofol general anesthetic for craniotomy be-fore awake functional brain mapping. Anesth Analg101: 502–508, 2005


7. Awake periodi) In principle, systemic administration of seda-tives and analgesics should not be done.ii) For light sedation, administer propofol, etc. atthe minimum required dose. (There are reports onthe use of dexmedetomidine as a sedative. Use ofremifentanil in patients with spontaneous respira-tion is not recommended because of respiratorydepression.)iii) If the patient complains of pain, provide addi-tional local anesthesia first.iv) If nausea and/or vomiting occur: Discontinuethe surgical procedure, administer metoclopramideor a serotonin receptor antagonist, and wait for thesubsidence of symptoms. Remove vomitus to pre-vent aspiration. If symptoms are severe and do notsubside, consider sedation with propofol and dis-cuss with the surgeons regarding the discontinua-tion of awake craniotomy.v) If convulsions develop: Discontinue the surgicalprocedure, especially electrical stimulation. (If theelectroencephalogram is being monitored, the oper-ation should be discontinued when the first spike isseen.) Cool the brain surface with cold water. Ad-

minister propofol at a sleeping dose. Give an in-travenous infusion of 250 mg of phenytoin. If con-vulsions do not cease even after additional adminis-tration of propofol, midazolam, or thiopental, dis-continue awake craniotomy.

[Commentary]During the awake period, as a general rule, sys-

temic application of sedatives or analgesics shouldnot be done in order to minimize the influence onfunctional mapping or the identification of epilepticfoci. To deal with pain, add local anesthetics. If asmall dose of a sedative or narcotic is considered toprevent worsening of the patient's mental state andexcitation, the potential influence on functional as-sessment should be assessed. Recently, there havebeen several reports about anesthesia during awakesurgery in which dexmedetomidine or remifentanilwas used during the awake period.1,3–6,9) However,there have also been reports that poor emergence ofpatients given dexmedetomidine required a de-crease of the dose or discontinuation, and dex-medetomidine is not covered by health insurance inJapan. The use of remifentanil at low doses withspontaneous respiration on emergence has also beenreported, but is not considered to be safe due to thepotential for respiratory depression or brain swell-ing induced by hypercapnia, so this method wouldrequire careful attention.

Although the incidence of nausea and vomitingduring awake surgery varies among reports, it hasbeen reported to be approximately 0–10% whenanesthetic management is primarily done withpropofol.8) Nausea and vomiting, in addition to caus-ing discomfort for the patient, increases the risk ofrespiratory complications due to aspiration, andbody movement and increased brain swelling asso-ciated with nausea/vomiting may make the surgicalprocedure more difficult. Nausea and vomiting maybe induced by the surgical procedure or by use ofnarcotics. At the onset, immediately discontinue thesurgical procedure and administer metoclopramideor a serotonin receptor antagonist. However, seroto-nin receptor antagonists are only available off labelin Japan, requiring the decision to be made at eachinstitution. If symptoms are severe and do not im-prove, consider sedation with propofol and evenconsider the discontinuation of awake craniotomyin certain cases. Although there are some reportsabout medications to prevent nausea and vomiting,the efficacy during awake surgery is unknown.

The incidence of convulsions during awakecraniotomy depends on the underlying disease andis reported to be approximately 0–24%.2,8) Convul-sions are more likely to develop during electrical

137137



stimulation for brain functional mapping. If convul-sions develop, discontinue electrical stimulationduring the surgical procedure and cool the brainwith cold Ringer's solution or saline. If the elec-troencephalogram is being monitored, discontinuethe procedure at the onset of a spike. Most convul-sions cease with discontinuation of the surgicalprocedure and cooling of the brain. If these meas-ures are ineffective, administer propofol orphenytoin at a sleeping dose. The preventative effectof phenytoin has not been confirmed, so it is consi-dered desirable to achieve an effective blood con-centration before surgery. If convulsions do notcease with additional propofol, midazolam, orthiopental, discontinue awake craniotomy. Therehas been a report that intractable convulsions re-quired general anesthesia with tracheal intubation.7)

During awake craniotomy, it is necessary to be pre-pared for emergency transition to airway manage-ment or general anesthesia at any time.

References1) Ard JL Jr, Bekker AY, Doyle WK: Dexmedetomidine in

awake craniotomy: a technical note. Surg Neurol 63:114–116, 2005

2) Conte V, Baratta P, Tomaselli P, Songa V, Magni L,Stocchetti N: Awake neurosurgery: an update. Miner-va Anesthesiol 74: 289–292, 2008

3) Herrick IA, Craen RA, Blume WT, Novick T, GelbAW: Sedative doses of remifentanil have minimal ef-fect on ECoG spike activity during awake epilepsy sur-gery. J Neurosurg Anesthesiol 14: 55–58, 2002

4) Lobo F, Beiras A: Propofol and remifentanil effect-siteconcentrations estimated by pharmacokinetic simula-tion and bispectral index monitoring duringcraniotomy with intraoperative awakening for braintumor resection. J Neurosurg Anesthesiol 19: 183–189,2007

5) Mack PF, Perrine K, Kobylarz E, Schwartz TH, LienCA: Dexmedetomidine and neurocognitive testing inawake craniotomy. J Neurosurg Anesthesiol 16: 20–25,2004

6) Sarang A, Dinsmore J: Anaesthesia for awake cranio-tomy—evolution of a technique that facilitates awakeneurological testing. Br J Anaesth 90: 161–165, 2003

7) Serletis D, Bernstein M: Prospective study of awakecraniotomy used routinely and nonselectively forsupratentorial tumors. J Neurosurg 107: 1–6, 2007

8) Skucas AP, Artru AA: Anesthetic complications ofawake craniotomies for epilepsy surgery. AnesthAnalg 102: 882–887, 2006

9) Souter MJ, Rozet I, Ojemann JG, Souter KJ, HolmesMD, Lee L, Lam AM: Dexmedetomidine sedation dur-ing awake craniotomy for seizure resection: effects onelectrocorticography. J Neurosurg Anesthesiol 19:38–44, 2007

8. Reinduction and completion of craniotomyi) When the cooperation of the patient is not re-quired any further, induce sedation with propofol.ii) In principle, manage the patient with spontane-ous respiration. However, if the airway needs to besecured because of oversedation, use the LMA.(Anesthesiologists who are experienced in handlingthe LMA may perform anesthetic management bydeliberately using it at closure.)iii) If needed, add more local anesthesia. However,if there is evidence of local anesthetic toxicity, dis-continue additional anesthesia and provide necessa-ry treatment such as establishment of an airway andcountermeasures for convulsions.iv) If the airway is established with an LMA, the re-quired dose of fentanyl or remifentanil can be givenfor analgesia.

[Commentary]Propofol is generally used as the anesthetic at the

end of craniotomy, as it is during craniotomy.1)

Determine whether tumor resection will be per-formed in the awake state or under sedation withpropofol, considering the conditions at each institu-tion and each patient. Some surgeons want patientsto be re-awakened after tumor resection to check forneurological symptoms.1) Insertion of an LMAshould be done via a lateral caudal approach and re-quires some degree of proficiency when the head isfixed with pins. There is a risk of difficulty with air-way establishment or vomiting and is recommendedfor at least two anesthesiologists to be involved in in-serting the LMA. After establishment of the airwaywith an LMA, management can be achieved by con-trolled respiration with remifentanil or fentanyl. Ifestablishment of the airway takes a long time,tracheal intubation2) can be considered. If establish-ment of the airway is not done, add further localanesthetic to continue surgery. Preparations shouldbe made to allow for establishment of the airwaywith an LMA immediately after a sudden change ofthe patient's state, such as the onset of convulsions.If analgesia is insufficient, add a small dose of fen-tanyl. Caution is required with regard to the use ofremifentanil with spontaneous respiration duringcraniotomy, as on emergence.

References1) Sato K, Kawamata M, Nagata O, Kawaguchi M,

Morimoto Y, Kato M, Sakabe T: [Present state ofanesthetic management for awake craniotomy inJapan]. Masui 57: 492–496, 2008 (Japanese)

2) Suzuki A, Terao M, Aizawa K, Yamagishi A, Kurosa-wa A, Iwasaki H: [Experience of the use of airwayscope and airtrack in a face-to-face sitting position].Rinsho Masui 32: 1327–1334, 2008 (Japanese)

138138



9. Emergence and dischargei) After the completion of surgery, discontinuesedatives and analgesics.ii) Confirm emergence and the return of spon-taneous respiration, remove the LMA if used, andtransfer the patient to the intensive care unit.

[Commentary]Follow the regular procedures for neurosurgical

anesthetic management of emergence and dis-charge.

III. LANGUAGE ASSESSMENT DURING AWAKE CRANIOTOMY

1. Methods of language mapping by cortical elec-trical stimulation during awake craniotomy[Recommendation]

Indications: Patients with lesions around theperisylvian language areas of the dominanthemisphere. Patients without apparent aphasia whoare able to fully understand the language tasks andcooperate with them.

Preoperative preparation: Set language tasks thatcan easily be performed by patients and fullyfamiliarize them with the tasks.

Electrical stimulation: A stronger stimulus inten-sity (6–12 mA) and longer duration (2–4 sec) are re-quired than for motor and sensory mapping. Initiateelectrical stimulation immediately before presentingthe language stimulus (line drawing or question) andcontinue it during presentation.

Language tasks: Perform counting, visual naming,and listening comprehension tasks for cortical map-ping. If electrical stimulation reveals any dysfunc-tion, assess reproducibility. Monitor language func-tions primarily on the basis of spontaneous speechduring resection. If an abnormality is suspected,perform language mapping with visual namingand/or listening comprehension.

[Commentary]Purpose: The purpose of language mapping is to

identify the language areas and to avoid postopera-tive aphasia by preserving these areas. Because theextent of the language areas varies among individ-uals and it is difficult to accurately identify themanatomically, the areas should be determined foreach individual.5) If language areas are identifiedoutside the resection zone and language functionsare confirmed to be localized away from the lesionto be resected, the neurosurgeon can resect the le-sion with confidence.

Indications: Because patients must have full un-derstanding and good cooperation to perform lan-guage mapping, we should consider the preopera-tive cognitive level and mental maturity of thepatient. Because some patients cannot adapt to thespecial circumstances of the operating room en-vironment, after providing sufficient explanation

and practice of the tasks, determine whether theyare suitable candidates for awake surgery or not. Beespecially careful with young and elderly persons.Children and patients with obvious aphasia beforesurgery are not suitable for language mapping.Patients showing slight anomia or word findingdifficulty (poor word production by category or ini-tial phonemes) during preoperative examinationmay attempt language mapping, but their languagefunction can become worse than preoperatively be-cause of drowsiness, which can make language map-ping difficult.

Preoperative preparation: Examination: Performneurological and neuropsychological testing. Ex-planation: Fully explain the language tasks that willbe used for mapping. Establishment of tasks: Afterperforming the standard tasks once, exclude stimulithat evoke unstable responses, leaving only thestimuli for which the patient can definitely providecorrect answers. Determine the rate of presentingline drawings at which the patient can answer com-fortably (2–5 sec intervals). Practice: The selectedtasks should be practiced several times until thepatient can answer with confidence. If examinationreveals suspected decrease of language function,perform the standard aphasia tests to measure theseverity of aphasia. Identification of language-relat-ed sites by fMRI might be useful to limit the area thathas to be explored by intraoperative mapping.6)

Electrical stimulation: Follow the usual standards,but remember that stronger and longer stimulationis required for language mapping. Employ a stimu-lus intensity of 6–12 mA if no afterdischarge isevoked. Because electrical stimulation is initiatedimmediately before presenting a line drawing orquestion and is continued during presentation,stimulation needs to be applied for between 3 and 4seconds (Fig. 5). It is desirable to present the lan-guage tasks at regular intervals so that the neurosur-geons get used to the timing of electrical stimulation.Also, adjust the location of the screen so the sur-geons are able to monitor stimulation during the lan-guage tasks.

Language tasks: Language tasks to be performed:For all areas to be tested, counting and visual nam-

139

Fig. 5 Flow chart of language tasks. Initiate electrical stimulation before presenting the language tasks.

139



ing should be performed. With counting, checkspeech arrest and delay. Confirm that the site ofspeech arrest does not correspond to the negativemotor area. During the visual naming task (picturenaming), record slips of the tongue (errors), delayedresponses, or no response. Words for the namingtask are selected from among high-frequency words,such as cat, knife, desk etc. For the temporal lobe,also perform auditory comprehension. Frequency:Stimulate each site twice or more with the maximumcurrent to check whether a language abnormality isdetected. If any language abnormality is detected,stimulate the site twice or more again to check re-producibility. Interpretation of results: If threestimuli induce at least two incorrect responses, thesite should be designated as a language-related site(Fig. 5).

Cortical mapping: Counting (from 1 to 30): Per-form electrical stimulation while asking the patientto count from 1 to 30 at approximately one numberper second. After the patient reaches 30, he/shestarts from 1 again. Identify the sites where stimula-tion leads to abnormalities of speech (arrest, delay,dysarthria). Regarding the sites associated withthese abnormalities, ask about the patient's subjec-tive symptoms (e.g., inability to move the tongue).Then, assess whether or not the sites are primarymotor areas or negative motor areas related to ar-ticulation. Visual naming7): Present line drawings(on paper or a monitor) at the interval predeter-mined for each patient and instruct the patient toname them using a carrier phrase like ``This is...''.Anomia or paraphasia: After saying ``This is'' fluent-

ly, patients cannot recall the name or substitute oneword for another. Speech arrest: The patient cannotsay ``This is''. Auditory comprehension: The patientanswers an easy question with a single word. Be-cause this involves both word recall and listeningcomprehension, electrical stimulation at sites differ-ent from those related to visual naming induces ab-normalities.3) Language mapping is based on theabove three tasks. If time permits, other languagetasks can be added as required.

Subcortical mapping: This is required if nervefibers immediately below or adjacent to the languageareas are to be resected. Continue conversation be-tween the patient and examiner during resectionand perform mapping with electrical stimulation atthe sites with possible abnormalities. Use visualnaming and, for the posterior language areas, listen-ing comprehension as well. The intensity of electri-cal stimulation should be equal to or slightly greaterthan that for cortical stimulation. Identification ofnerve fascicles by preoperative tractography mightbe useful to determine the sites for subcortical map-ping.1,2,4)

References1) Berman JI, Berger MS, Chung SW, Nagarajan SS, Hen-

ry RG: Accuracy of diffusion tensor magneticresonance imaging tractography assessed using in-traoperative subcortical stimulation mapping andmagnetic source imaging. J Neurosurg 107: 488–494,2007

2) Duffau H: The anatomo-functional connectivity of lan-guage revisited. New insights provided by elec-trostimulation and tractography. Neuropsychologia

140

Table 1 Examples of line drawings used in the visual naming task

Grape Ear Ant Potato Train Strawberry Eye Cat Truck RabbitBus Scissors Patrol car Carrot Plane Chicken Pencil Motorcycle Apple Cup

Words are selected from among high-frequency words in the vocabulary test for aphasia and color drawings withoutcopyright are used.

Table 2 Examples of questions used to test auditorycomprehension

1. What is your name?2. What color is snow?3. What color is a sunflower?4. What color is a crow?5. What color is a banana?6. What color is a fire truck?7. How many days are there in a week?8. How many minutes are there in an hour?9. How many legs has a dog?

10. What day is after Tuesday?11. What season is after spring?12. What month is New Year's Day in?13. What month is the Bon Festival in?14. Which direction the sun sets in?15. What is the offspring of a frog called?16. What is the offspring of a chicken called?17. A mother is a woman. What is a father?18. A brother is a man. What is a sister?19. The sun shines during the daytime. When do the stars come

out?20. Cherry blossom is seen in spring. How about red leaves?21. Where do you buy postage stamps?22. What do you use to cut vegetables?23. What do you use to cut paper?24. What do you use to tell the time?25. Hot water is hot. How about ice?26. Iron is heavy. How about feathers?27. The sea is deep. How about mountains?28. You wear clothes. How about shoes?29. Birds fly. How about fish?30. You listen to music. How about paintings?

Questions that can be answered within approximately 2seconds are prepared based on the Wechsler IntelligenceScale for Children-Revised, Illinois Test of Psycholinguis-tic Abilities, Western Aphasia Battery test, etc.

140



46: 927–934, 20083) Hamberger MJ, Seidel WT: Auditory and visual nam-

ing tests: normative and patient data for accuracy,response time, and tip-of-the-tongue. J Int Neuropsy-chol Soc 9: 479–489, 2003

4) Henry RG, Berman JI, Nagarajan SS, Mukherjee P,Berger MS: Subcortical pathways serving cortical lan-guage sites: initial experience with diffusion tensorimaging fiber tracking combined with intraoperativelanguage mapping. Neuroimage 21: 616–622, 2004

5) Ojemann G, Ojemann J, Lettich E, Berger M: Corticallanguage localization in left, dominant hemisphere.An electrical stimulation mapping investigation in 117patients. J Neurosurg 108: 411–421, 2008

6) Rutten GJ, Ramsey NF, van Rijen PC, Noordmans HJ,van Veelen CW: Development of a functional magne-tic resonance imaging protocol for intraoperativelocalization of critical temporoparietal languageareas. Ann Neurol 51: 350–360, 2002

7) Sanai N, Mirzadeh Z, Berger MS: Functional outcomeafter language mapping for glioma resection. N Engl JMed 358: 18–27, 2008

2. Task detailsExamples of line drawings used in the visual nam-

ing task and questions used to treat auditory com-prehension are shown in Tables 1 and 2, respec-tively.

141141



Appendix: GUIDELINES COMMITTEE OF THE JAPAN AWAKE SURGERY CONFERENCE

Chairman of the CommitteeTakamasa KAYAMA (National Cancer Center & Dept. of Neurosurgery, Yamagata University Faculty of Medicine)

Co-Chairman of the CommitteeHiroshi ISEKI (Tokyo Women's Medical University, Faculty of Advanced Techno-Surgery)Yoshitsugu YAMADA (Dept. of Anesthesiology, Faculty of Vital Care Medicine, The Graduate School of Medicine, The

University of Tokyo)

Committee MembersTatsuya ABE (Dept. of Neurosurgery, Oita University Faculty of Medicine)Chikashi FUKAYA (Dept. of Neurosurgery and Applied System Neuroscience, Nihon University School of Medicine)Hidekatsu FURUTANI (Dept. of Anesthesia, Kyoto University Hospital)Kazuhiro HONGO (Dept. of Neurosurgery, Shinshu University School of Medicine)Tohru ITAKURA (Wakayama Medical University)Koji KAJIWARA (Dept. of Neurosurgery, Yamaguchi University Graduate School of Medicine)Kyosuke KAMADA (Dept. of Neurosurgery, Asahikawa Medical University)Masahiko KAWAGUCHI (Dept. of Anesthesiology, Nara Medical University)Mikito KAWAMATA (Dept. of Anesthesiology and Resuscitology, Shinshu University School of Medicine)Toshihiro KUMABE (Dept. of Neurosurgery, Tohoku University Graduate School of Medicine)Junichi KURATSU (Dept. of Neurosurgery, Kumamoto University School of Medicine)Masanori KURIMOTO (Dept. of Neurosurgery, University of Toyama, Faculty of Medicine)Nobuhiro MIKUNI (Dept. of Neurosurgery, Sapporo Medical University, School of Medicine)Yasuhiro MORIMOTO (Dept. of Anesthesiology, Yamaguchi University Graduate School of Medicine)Yoshihiro MURAGAKI (Tokyo Women's Medical University, Faculty of Advanced Techno-Surgery)Osamu NAGATA (Dept. of Anesthesiology, Tokyo Women's Medical University)Kaori SAKURADA (Dept. of Neurosurgery, Yamagata University Faculty of Medicine)Kiyotaka SATO (Dept. of Anesthesiology, Kohnan Hospital)Shinya SATO (Dept. of Neurosurgery, Yamagata University Faculty of Medicine)Nobusada SHINOURA (Dept. of Neurosurgery, Komagome Metropolitan Hospital)Morihiro SUGISHITA (Institute of Brain Vessels)Masahiko SUMITANI (Dept. of Anesthesiology and Pain Relief Center, The University of Tokyo)Kyoko SUZUKI (Dept. of Clinical Neuroscience, Yamagata University Faculty of Medicine)Seiji TAKAOKA (Dept. of Anesthesiology, Yamagata City Hospital Saiseikan)

The Guidelines for Awake Craniotomy

Documents

Transcript of The Guidelines for Awake Craniotomy