Precise intra-operative localization of functional areas is essential in order to avoid neurological deficits from microsurgical tumor removal in eloquent areas of the brain, especially in cases of invasive and malignant brain tumors. Even though there are well known landmarks to localize functional areas on the cortical surface, they are no longer applicable if functional areas and cerebral tracts are displaced by the space occupying lesion.
Awake craniotomies with intraoperative neuropsychological monitoring and cortical mapping have been recommended as the gold standard for tumor resections in or near eloquent cortical areas [4, 5, 9, 10]. There is only one study in the literature that describes better results in patients operated under general anesthesia . The standard anesthesiological protocol for awake surgeries with cortical stimulation and language mapping is the asleep-awake-asleep method [17, 23, 29, 31], named after the three phases the patient is in during the procedure. The patient is “asleep” during the craniotomy, and then sedation is interrupted to allow the patient to be “awake” for cortical stimulation and language testing. After testing is completed, sedation is continued and the patient is again “asleep” for the closure of the skull and skin. Even though this protocol seems to be optimal for awake craniotomies as the patient is left asleep during the opening and closure of the skull, and is awoken only for language testing, it carries potential risks with respiratory complications and hemodynamic dysregulation being the most commonly reported [7, 17, 18, 23, 24, 29, 32–34]. Furthermore, using the asleep-awake-asleep protocol, patients are always more or less under the influence of the drugs from sedation, leaving them bradyphrenic during neuropsychological testing. This potentially falsifies functional test results creating uncertainty with regards to determining the exact borders of functional areas.
In an effort to eliminate potential risks for the patient during awake craniotomies, and in order to improve reliability of intra-operative test results that have a direct influence on the surgical outcome of awake craniotomies, we implemented a new protocol for awake craniotomies, the continuous awake craniotomy (CAC) protocol. In this protocol the patient is not asleep at any point as a result of sedation, hence the name continuous awake craniotomy protocol. We present the first clinical results of this new protocol.
Patients and methods
Patients and data
In this prospective study data of 13 awake craniotomies, performed between September 2006 and June 2008, according to the CAC protocol were analyzed. The indications to perform an awake craniotomy were primary brain tumors in or close to a speech center, their connecting cerebral tracts, or the sensorimotor cortex. The Karnofsky score had to be 70% or higher for patients to be considered for an awake surgery. Data used for analyses included OR-records, anesthesiological protocols, patients’ charts, and neuropsychological records. All patients gave written consent to undergo awake craniotomies and the study was also approved by the local ethics committee.
To get the essential structural and functional neuroanatomic information for precise planning of an awake craniotomy, the following imaging protocol was used. Morphological T1, 3D MP-RAGE magnetic resonance imaging (MRI) with contrast enhancement before and within 48 h after surgery as well as T2 sequences, blood oxygen level dependent (BOLD) functional MRIs, and diffusion tensor imaging (DTI) were acquired. A quantitative tumor volume analysis was performed in all cases before and after surgery. A gross total resection (GTR) was defined as a tumor volume reduction of 100% based on volumetric measurements in the postoperative MRI with contrast enhancement and in cases of a low grade glioma (LGG), the volume of the flair in the T2 flair images.
Brain mapping and neuropsychological evaluation
Extensive brain mapping with direct cortical stimulation was performed in all patients before tumor resections were started. Neuropsychological testing and cortical stimulation during glioma surgeries is important to allow the location of functional areas, as shown on the fMRI scans, to be confirmed before tumor resection is started (see case report). As gliomas show an invasive pattern of growths, it is essential to detect functional borders without compromising the radicality of the resection by stopping too soon. A GTR should always be the aim during resections of gliomas. If the tumor is located in or near a functional speech area, neuropsychological testing in combination with subcortical stimulation during the resection helps to detect functional borders and fiber tracts before permanent neurological deficits are caused.
For the aforementioned reasons, subcortical stimulation was repeatedly performed under neuropsychological testing during tumor resections. Out of a range of available standardized tests, adapted for German speaking people, an individualized test battery adapted to the intellectual level was put together for each patient. Basic neuropsychological tests for pre-, intra- and postoperative evaluation of language function included assessment of lexical fluency, semantic fluency, similarities (conceptualization) verbal and nonverbal cognition as well as verbal intelligence [1, 3, 8]. Evaluation of motor function was performed with the Tapping Board , the Dynamometer  as well as the Grooved Pegboard . Intra-operatively, only the Tapping Board was used.
Neuroanesthesia: the continuous awake craniotomy protocol
The introduced protocol is called continuous awake craniotomy protocol since contrary to the asleep-awake-asleep protocol the patient is awake during the entire procedure and there are no sedatives used.
All patients received an intravenous line as well as arterial cannulation (radial artery) for continuous arterial blood pressure monitoring and blood gas analyses. Non-invasive monitoring of vital functions included electrocardiography, a blood pressure cuff, pulsoximetry, and a nasal oxygen mask. Respiratory rate and end-tidal carbon dioxide were measured using a nasal prongs-port with capnometry (CO2 port). All patients received a urinary catheter. No airway instrumentation was used, however, this was kept within reach of the anesthesiologist in case of a non-self-limiting seizure causing a drop in oxygen saturation (<90%) occurring.
Using 0.75% Ropivacain (AstraZeneca GmbH, Wedel, Germany) supplemented with Adrenalin (AstraZeneca GmbH) diluted 1/200,000 scalp nerve blocks (SNB) were placed through a 23 gauge needle using a previously published method . SNB were placed at least 20 min before the Mayfield head-clamp was put on.
A close and individual intraoperative patient guidance and pacing, SNB supplemented with Remifentanil-analgesia (GlaxoSmithKline GmbH, Munich, Germany) whenever needed were used for all patients. Patients did not receive sedative agents during the entire procedure except in the event of a generalized seizure not controllable by irrigation of the brain with cold ringer solution [26, 30]. Anticonvulsive medication using Levetiracetam (Keppra; UCB GmbH, Monheim am Rhein, Germany) was used for two patients with a history of seizures and who were already on anticonvulsive medication.
Data of 12 consecutive patients (six men/six women) with a median age of 45.5 years (range 28–70) who underwent a total of 13 awake craniotomies using the CAC protocol were included in this study. SNBs were tolerated well by all patients and the effect of local anesthesia lasted during the entire surgery (range 129–306 min) in all cases and did not have to be renewed in any of the patients. Remifentanil-analgesia was given in a mean total dosage of 374 µg with 0.1–0.4 mg/h, and over a mean accumulated time period of 60 min (range 50–85 min), and showed no relevant sedative effects.
Two patients suffered seizures (one focal/one generalized) intra-operatively triggered by direct cortical stimulation. In case of the generalized seizure, surgery had to be continued under sedation as the seizure could not be controlled by applying cold Ringer solution on the cortical surface. However, this patient did not have to be intubated. Nevertheless, this case counted as an awake craniotomy with respect to the surgical outcome. A focal seizure was self-limited, short-lived and could be controlled by irrigation of the brain with cold Ringer solution. None of the other surgeries had to be stopped due to patient fatigue, respiratory problems, swelling of the brain, or excessive bleeding. No blood pressure problems were observed during any of the procedures. None of the patients complained of excessive pain, discomfort or suffered from nausea and vomiting intra-operatively or during the postoperative phase. The CAC protocol was tolerated well by all patients. This was described in more detail in another publication from our group . The median duration of surgery (skin to skin) was 215 min (range 135–300 min) and median testing time (speech and/or motor function) with active participation of the patient during tumor resections was 155 min (range 75–240 min).
Surgical and functional outcome
The median tumor volume was 14.2 cm3 (range 0.5–45.7 cm3). Two patients with brain metastasis in eloquent areas were also included in the study as preoperative imaging showed no clear borders between the tumor and healthy brain tissue. A GTR determined on early postoperative MRI was achieved in ten patients (83.3%).
One patient (8.3%) (Table 1, number 7) suffered from a new permanent neurological deficit postoperatively. In the immediate postoperative phase one other patient (Table 1, number 12) suffered from a temporary accentuation of a hemiparesis that was already present preoperatively. The patient whose surgery was continued under sedation showed a paresis of the left arm immediately after surgery, however, this was only temporary. The mean Karnofsky score was 94% before and 90% after surgery. Presenting symptoms, histological results, tumor locations, tumor volumes as well as achieved extend of resection are summarized in Table 1.
Diagnosis and patient preparation
One of the patients treated in this study was a 44-year-old right handed male patient (Table 1, number 2) with a left fronto-lateral glioblastoma. The initial symptoms were a sudden speech arrest followed by a generalized seizure which occurred after the patient went for a sauna 4 weeks before being admitted to our department. After being brought to a hospital the initial cranial computed tomography (CT) showed a right fronto-lateral lesion. An MRI was then performed showing a lesion suspicious of a glioblastoma. A biopsy was performed at another institution confirming the diagnosis of a glioblastoma. After being admitted to our institution for further treatment, the patient underwent thorough preoperative imaging which included T1, 3D MP-RAGE MRI with contrast enhancement (Figure 1), T2 sequences, DTI as well as BOLD functional MRI scans. Functional neuroradiological imaging confirmed the lateralization of the language mainly on the right side next to the contrast enhancing lesion (Figure 1). Based on this finding, it was decided that an awake craniotomy was the best strategy to preserve the language function. The patient gave his written informed consent for an awake surgery in the CAC protocol. Before surgery the neuropsychologist performed several tests that were selected from a range of available standardized tests adapted for German speaking people. The same test battery was then also used for the intra-, and postoperative assessment.
The patient received an arterial cannulation (radial artery) for continuous blood pressure monitoring and blood gas analyses. Non-invasive monitoring of vital functions included electrocardiography, a blood pressure cuff and pulsoximetry. He also received a nasal oxygen mask with capnometry (CO2 port) and a urinary catheter. Using 0.75% Ropivacain (AstraZeneca GmbH) supplemented with Adrenalin (AstraZeneca GmbH) diluted 1/200,000, the scalp nerve blocks were placed through a 23 gauge needle to block the greater occipital nerves (4 mL on each side), lesser occipital nerves (3 mL on each side), auriculotemporal nerves (5 mL on each side), supratrochlear nerves (1 mL on each side), supra-orbital nerves ( 2 mL on each side), and the zygomaticotemporal nerves (2 mL on each side). Then the patient was positioned on his left side, the Mayfield clamp was put on and the surgery was started.
During the craniotomy, the patient was closely guided by the anesthesiologist using the method of care giving and accompanying patient communication. After performing an individually tailored craniotomy without any sedation, the transition to the neuropsychological testing was seamless.
Intraoperative cortical mapping was performed and showed functional tissue in the area of the planned surgical approach, and not only anterior to the lesion as the fMRI suggested (Figures 1 and 2). The only part with “non-functional” tissue was found around the scar tissue in the area of the previously performed biopsy (Figure 2). The corridor to the tumor created by the biopsy needle was used during the awake tumor resection allowing a very small and direct approach to the tumor (Figure 3). The tumor resection was performed under permanent neuropsychological testing and repeated subcortical stimulations. Total surgery time (skin to skin) was 306 min and there were no cardiovascular or respiratory complications. The patient was alert at all times and had no problems during neuropsychological testing.
Surgical and functional outcome
A postoperative MRI (Figure 4) showed that a gross total resection was achieved. The patient had no new neurological deficits after surgery and speech function was unaffected. After surgery the patient underwent an adjuvant treatment with radiation and chemotherapy.
The presented CAC anesthesia protocol omits any sedation during awake surgery and in this way it is different from the commonly used “asleep-awake-asleep” protocol . The present study shows the feasibility of this novel protocol as it allows for a great radicality during surgery, and can help to decrease the operative morbidity. Though all patients suffered from a tumor in the close vicinity of functional relevant brain areas, a new neurological deficit was observed in only one of the twelve cases included in this study.
A review of the recent literature (Table 2) revealed that the asleep-awake-asleep protocol and the method of conscious sedation of a patient are most commonly used during awake craniotomies [2, 11, 13, 16, 18, 22–24, 27–29, 32–34]. Interestingly, similar to the CAC protocol presented here, some authors report the additional use of skull nerve blocks for regional pain management during opening of the skin and the craniotomy [31, 33]. While there are no generally accepted standards, the patient is usually mildly sedated, agents such as Propofol and Fentanyl are used for analgesia [2, 13, 23, 29, 34]. Studies revealed major drawbacks of that protocol, such as respiratory complications in up to 37% [18, 23, 24, 29, 32, 33], circulatory dysregulation in up to 22% [23, 24, 29, 32, 33], brain edema in up to 14% [23, 24, 27, 33], and uncooperative or agitated patients in up to 20% [18, 23, 24, 27, 29]. Management of the airways has been recognized as a particularly critical part of the standard awake-asleep-awake protocol with the danger of aspiration due to nausea and vomiting in the phase where the patient wakes up during the surgery for language testing. Various techniques of the airway management are described in the literature, ranging from leaving the patient breathing spontaneously  and securing the airway transnasally with a Magill tube  or a laryngeal mask . In some cases, patients are even ventilated using a laryngeal mask  or intubated for the asleep part of the surgery . We believe that airway related complications are caused by the use of sedative agents during the “asleep” phase of awake craniotomies. It is also obvious that these substances can influence the practicability and validity of neuropsychological testing during the surgery. Objectively, opioids could also have a sedative effect, and therefore our method is not convincing. However, the sedative effect of opioids in the low dosages as administered in our protocol, must be considered to be negligible. On that note, several studies specifically analyzing and evaluating the effects of drugs on human memory [14, 15] have shown that general anesthetics do impair short term memory, which is highly relevant for intra-operative language and memory testing, while opioids do not show such an effect . Even though optimized protocols for the asleep-awake-asleep method have been introduced [6, 27, 33], the timing of when to reduce sedation remains difficult and sedation related side effects cannot always be prevented.
Continuous awake craniotomy protocol
It is the purpose of the presented CAC protocol to address the mentioned issues of the asleep-awake-asleep protocol just by avoiding the sedative medication. Avoiding the wake-up phase and omitting the sedation should reduce the physiological stress of the patient from anesthesia. Wake-up stress related complications can result from sympathetic activation, catecholamine release, and may be accompanied by an increase in blood pressure or heart rate as well as increased cerebral blood flow and cerebral oxygen consumption .Also, an increase in PaCO2 during the wake-up phase may result in cerebral swelling changing the threshold for cortical stimulation and by that introducing yet another factor that creates uncertainty . Those physiological reactions may contribute to situations where patients become agitated in the wake-up phase [18, 24, 29, 34], which potentially may result in an uncontrollable situation. Avoiding the wake-up phase, our results indicate that our protocol lacks such anesthesia related arousal states.
However, despite the great advantages of the CAC protocol, the risk of causing a seizure by cortical stimulation remains [27, 30]. However, this is the only real non-surgery related risk during an awake craniotomy in the CAC protocol. In order to make results better comparable, we quantified our intra-operative test results, however, we did not find any other publications to compare our results to as most authors only vaguely describe their intra-operative test results [18, 23, 24, 27, 32]. In spite of extensive testing during tumor resection (range 75–240 min), none of the patient testing had to be stopped as a result of fatigue of the patient as reported by other groups .
From a practical point of view, the CAC protocol is advantageous, as it allows the neurosurgeon to start the testing procedure at any time, on demand without having to coordinate the wake-up phase of the patient. It allows the neurosurgeon to perform testing repeatedly during surgery and not only, as it is frequently practiced with the asleep-awake-asleep protocol, for a single testing at the beginning of surgery in order to determine the anatomical relationship of the lesion and the functional area. Considering the mentioned potential risks of the asleep-awake-asleep protocol [24, 29, 32, 33], our results, even in a small cohort of patients, indicate significant advantages of the CAC protocol. None of our patients had respiratory complications or showed hemodynamic dysregulation. All patients performed very well during neuropsychological and electrophysiological testing. Based on volumetric evaluations, a gross total resection was achieved in ten out of 12 patients (83%), which is better than results reported by many other groups [2, 6, 16, 22, 28] where GTRs range between 32% and 77% (Table 2), only Picht et al. reported 100% GTR .
In conclusion, our results indicate a possible advantage of the novel CAC protocol. The results from our pilot study warrant a prospective randomized clinical trial comparing the CAC with the asleep-awake-asleep protocol.
Aschenbrenner S, Tolk W. Regensburger Wortflüssigkeitstest RWT Handanweisung. Göttingen: Hogrefe; 2000.Google Scholar
Bello L, Gallucci M, Fava M, Carrabba G, Giussani C, Acerbi F, et al. Intraoperative subcortical language tract mapping guides surgical removal of gliomas involving speech areas. Neurosurgery. 2007;60:67–80.PubMedWeb of ScienceGoogle Scholar
Benton AL, Hamsher K. Multilingual aphasia examination. Iowa City: AJA Associates; 1989.Google Scholar
Berger MS, Kincaid J, Ojemann GA, Lettich E. Brain mapping techniques to maximize resection, safety, and seizure control in children with brain tumors. Neurosurgery. 1989;25:786–92.PubMedGoogle Scholar
Danks RA, Aglio LS, Gugino LD, Black PM. Craniotomy under local anesthesia and monitored conscious sedation for the resection of tumors involving eloquent cortex. J Neurooncol. 2000;49:131–9.CrossrefPubMedGoogle Scholar
Duffau H, Capelle L, Denvil D, Sichez N, Gatignol P, Taillandier L, et al. Usefulness of intraoperative electrical subcortical mapping during surgery for low-grade gliomas located within eloquent brain regions: functional results in a consecutive series of 103 patients. J Neurosurg. 2003;98:764–78.CrossrefGoogle Scholar
Duffau H, Peggy Gatignol ST, Mandonnet E, Capelle L, Taillandier L. Intraoperative subcortical stimulation mapping of language pathways in a consecutive series of 115 patients with Grade II glioma in the left dominant hemisphere. J Neurosurg. 2008;109:461–71.CrossrefWeb of ScienceGoogle Scholar
Fukaya C, Katayama Y, Yoshino A, Kobayashi K, Kasai M, Yamamoto T. Intraoperative wake-up procedure with propofol and laryngeal mask for optimal excision of brain tumour in eloquent areas. J Clin Neurosci. 2001;8:253–5.CrossrefPubMedGoogle Scholar
Gupta DK, Chandra PS, Ojha BK, Sharma BS, Mahapatra AK, Mehta VS. Awake craniotomy versus surgery under general anesthesia for resection of intrinsic lesions of eloquent cortex–a prospective randomised study. Clin Neurol Neurosurg. 2007;109:335–43.CrossrefPubMedWeb of ScienceGoogle Scholar
Keifer JC, Dentchev D, Little K, Warner DS, Friedman AH, Borel CO. A retrospective analysis of a remifentanil/propofol general anesthetic for craniotomy before awake functional brain mapping. Anesth Analg. 2005;101:502–8.CrossrefPubMedGoogle Scholar
Lafayette Instrument. Hand dynamometer instruction/owner’s manual. Lafayette: Lafayette Instrument Company; 1986.Google Scholar
Lafayette Instrument. Grooved pegboard test instruction/owner’s manual. Lafayette: Lafayette Instrument Company; 1989.Google Scholar
Lezak MD. Neuropsychological assessment. New York: Oxford University Press; 2004.Google Scholar
Low D, Ng I, Ng WH. Awake craniotomy under local anaesthesia and monitored conscious sedation for resection of brain tumours in eloquent cortex–outcomes in 20 patients. Ann Acad Med Singapore. 2007;36:326–31.PubMedGoogle Scholar
Manninen PH, Balki M, Lukitto K, Bernstein M. Patient satisfaction with awake craniotomy for tumor surgery: a comparison of remifentanil and fentanyl in conjunction with propofol. Anesth Analg. 2006;102:237–42.CrossrefPubMedGoogle Scholar
Milian M, Luerding R, Ploppa A, Decker K, Psaras T, Tatagiba M, et al. “Imagine your neighbor mows the lawn”: a pilot study of psychological sequelae due to awake craniotomy. J Neurosurg. 2013; [Epub ahead of print]. DOI: 10.3171/2013.2.JNS121254.CrossrefWeb of ScienceGoogle Scholar
Picht T, Kombos T, Gramm HJ, Brock M, Suess O. Multimodal protocol for awake craniotomy in language cortex tumour surgery. Acta Neurochir (Wien). 2006;148:127–37; discussion 137–8.CrossrefGoogle Scholar
Pinsker MO, Nabavi A, Mehdorn HM. Neuronavigation and resection of lesions located in eloquent brain areas under local anesthesia and neuropsychological-neurophysiological monitoring. Minim Invasive Neurosurg. 2007;50:281–4.PubMedWeb of ScienceGoogle Scholar
Sartorius CJ, Berger MS. Rapid termination of intraoperative stimulation-evoked seizures with application of cold Ringer’s lactate to the cortex. Technical note. J Neurosurg. 1998;88:349–51.CrossrefGoogle Scholar
The authors stated that there are no conflicts of interest regarding the publication of this article.
About the article
Published Online: 2013-05-31
Published in Print: 2013-06-01