- Email: [email protected]
Awake language mapping and 3-Tesla intraoperative MRI-guided volumetric resection for gliomas in language areas
Journal of Clinical Neuroscience 20 (2013) 1280–1287
Contents lists available at SciVerse ScienceDirect
Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn
Clinical Study
Awake language mapping and 3-Tesla intraoperative MRI-guided volumetric resection for gliomas in language areas Junfeng Lu a, Jinsong Wu a,⇑, Chengjun Yao a, Dongxiao Zhuang a, Tianming Qiu a, Xiaobing Hu b, Jie Zhang a, Xiu Gong a, Weimin Liang b, Ying Mao a, Liangfu Zhou a a b
Division of Glioma Surgery, Department of Neurosurgery, Huashan Hospital, Shanghai Medical College, Fudan University, Shanghai 200040, China Department of Anesthesiology, Huashan Hospital, Shanghai Medical College, Fudan University, Shanghai, China
a r t i c l e
i n f o
Article history: Received 14 August 2012 Accepted 7 October 2012
Keywords: Awake surgery Direct electrical stimulation Glioma Intraoperative MRI Volumetric analysis
a b s t r a c t The use of both awake surgery and intraoperative MRI (iMRI) has been reported to optimize the maximal safe resection of gliomas. However, there has been little research into combining these two demanding procedures. We report our unique experience with, and methodology of, awake surgery in a movable iMRI system, and we quantitatively evaluate the contribution of the combination on the extent of resection (EOR) and functional outcome of patients with gliomas involving language areas. From March 2011 to November 2011, 30 consecutive patients who underwent awake surgery with iMRI guidance were prospectively investigated. The EOR was assessed by volumetric analysis. Language assessment was conducted before surgery and 1 week, 1 month, 3 months and 6 months after surgery using the Aphasia Battery of Chinese. Awake language mapping integrated with 3.0 Tesla iMRI was safely performed for all patients. An additional resection was conducted in 11 of 30 patients (36.7%) after iMRI. The median EOR significantly increased from 92.5% (range, 75.1–97.0%) to 100% (range, 92.6–100%) as a result of iMRI (p < 0.01). Gross total resection was achieved in 18 patients (60.0%), and in seven of those patients (23.3%), the gross total resection could be attributed to iMRI. A total of 12 patients (40.0%) suffered from transient language deficits; however, only one (3.3%) patient developed a permanent deficit. This study demonstrates the potential utility of combining awake craniotomy with iMRI; it is safe and reliable to perform awake surgery using a movable iMRI. Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction Gliomas are the most common primary brain tumors and are associated with a poor prognosis.1–3 The standard treatment for gliomas is tumor resection followed by adjuvant radiotherapy and chemotherapy. An increasing body of literature4–7 demonstrates that the extent of resection (EOR) is a predictor of survival, despite a lack of class I evidence.8–10 However, the aggressive resection of gliomas located in eloquent areas carries the risk of neurologic deficits that can lead to a loss of quality of life, which may subsequently affect survival.11 Thus, for neurosurgeons, the main challenge in glioma surgery is achieving the maximal tumor resection while still preserving eloquent areas. Various advanced techniques and methods, including neuronavigation, intraoperative MRI (iMRI), 5-aminolevulinic acid and electrocortical mapping, have facilitated the maximal safe resection of gliomas in eloquent areas.12–17 Image guidance using neuronavigation, which is based on preoperative imaging, has been used extensively for several ⇑ Corresponding author. Tel./fax: +86 215 288 8771. E-mail address: [email protected] (J. Wu). 0967-5868/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jocn.2012.10.042
decades to aid in the resection of intracranial tumors. However, intraoperative brain shift poses a limitation for conventional neuronavigation. Therefore, iMRI has attracted increasing interest in the past decade because it effectively detects tumor remnants and compensates for intraoperative brain deformation.18 However, with the application of iMRI alone, it is difficult to preserve neurologic function reliably and to delineate a real-time safe boundary for tumor resection, particularly in gliomas involving language areas. Currently, awake language mapping is generally accepted as the gold standard for language localization and has been used with success in several languages of the Indo-European family.14,19–21 However, the feasibility and reliability of this technique as applied to the Chinese population (that is, Sino-Tibetan family speakers), have seldom been evaluated quantitatively.22 In addition, although both iMRI and awake language mapping are independent and established techniques, few studies23–27 have reported their combination. Experience in combining these two complicated modalities is limited, and the evidence is not sufficiently robust to draw a conclusion regarding the effectiveness of this combination.
J. Lu et al. / Journal of Clinical Neuroscience 20 (2013) 1280–1287
In this prospective study, we describe our approach and experience with the integration of awake language mapping and movable 3.0 Tesla (T) iMRI for the resection of gliomas in language areas. We aimed to both increase the EOR and to decrease the postoperative morbidity by combining these techniques. 2. Patients and methods 2.1. Patients Patients with tumors located in language areas as indicated by preoperative MRI were considered for tumor resection aided by the combination of iMRI and awake language mapping. Patients were selected using the following inclusion criteria: (i) the patients were adults of Han ethnicity; (ii) they usually spoke Chinese in daily life; (iii) they were right-hand dominant; and (iv) cerebral gliomas were preoperatively suspected. The exclusion criteria included the following: (i) patients with contraindications to MRI scanning, intraoperative neurophysiologic monitoring, or awake craniotomy (including patients with pacemakers, obstructive sleep apnea or severe intracranial hypertension); and (ii) patients whose pathologic diagnoses were not gliomas. All of the patients underwent language functional assessment using the Aphasia Battery of Chinese,28 which is the Chinese standardized adaptation of the Western Aphasia Battery. The Aphasia Quotient (AQ) score (spontaneous speech, comprehension, repetition and naming) was used to measure language ability. The preoperative cognitive state was also assessed using the Mini Mental State Examination.24 Patients with severe language deficits (AQ < 50) or cognitive disorders (Mini Mental State Examination score < 23) were excluded.24 Written informed consent was obtained from all patients. This study was undertaken at the Huashan Hospital (Shanghai, China) with approval from the Huashan Institutional Review Board.
1281
guage areas were then reconstructed into 3D objects and superimposed onto the structural data (Fig. 1A–C) using a post-processing workstation (Syngo MultiModality Workplace, Siemens AG). The 3D relationships between the lesion and the language areas were provided to the neurosurgeons for decision making. On the day of surgery, the reconstructed series were imported into the surgical navigation system.
2.3. Awake craniotomy A monitored anesthesia care (MAC) approach was adopted for all patients. After the anesthesiologists administered premedication by infusing 0.02–0.03 mg/kg midazolam and 5 mg tropisetron, they prepared the patient with intravenous lines, a central venous catheter, an arterial line, and a urethral catheter. The supraorbital, supratrochlear, zygomaticotemporal, auriculotemporal, greater occipital, and lesser occipital nerves of both sides were then blocked using a mixture of lidocaine (0.67%) and ropivacaine (0.5%). Once the patient was brought into moderate sedation with boluses of intravenous propofol, the head was fixed into its fitted position using a custom-designed high-field MRI-safe head holder (DORO Radiolucent Headrest System, Pro Med Instruments GmbH, Freiburg, Germany), which was integrated with an IMRIS operating room table and head coils (IMRIS). Once the scalp was prepared and draped, remifentanil (0.01–0.03 lg/Kg/min) or dexmedetomidine (0.1–0.7 lg/Kg/h) was administered for analgesia. To minimize brain swelling, mannitol (1 g/kg) was intravenously infused before the dura was opened. A low dose of remifentanil (0.01 lg/ Kg/min) or dexmedetomidine (0.1 lg/Kg/h) was administered during mapping. Because of the minimal draping, as described below, no laryngeal mask airway or endotracheal tubing was applied during the iMRI acquisition phase in these patients. Throughout the operation, supplemental inspired oxygen was delivered by nasal cannula or facemask.
2.2. Preoperative examination and evaluation Cognitive status and language function were preoperatively assessed by a neuropsychologist as previously mentioned. The patients were diagnosed with language deficits if they scored below 93.8 on the AQ preoperatively.29 Preoperative consultations by neurosurgeons and anesthesiologists occurred on the day before the operation. Patients were trained by neurosurgeons to perform the tasks of counting, naming objects and reading single words. Anesthesiologists were in charge of the interpretation for intraoperative discomfort and cooperation. Preoperative brain images were obtained in the diagnostic room of an iMRI-integrated neurosurgical suite (IMRIS, Winnipeg, Manitoba, Canada) using a movable 3.0 T scanner (MAGNETOM Verio 3.0 T, Siemens AG, Erlangen, Germany) 1 day prior to surgical intervention. The imaging protocol included a contrast-enhanced threedimensional (3D) magnetization-prepared rapid-gradient echo (MPRAGE) sequence (time of repetition (TR), 1900 milliseconds; time of echo (TE), 2.93 milliseconds; flip angle, 9 degrees; field of view (FOV), 250 250 mm2; matrix size, 256 215; slice thickness, 1 mm; acquisition averages, 1) or a fluid-attenuated inversion recovery (FLAIR) sequence (TR, 9000 milliseconds; TE, 96 milliseconds; TI, 2500 milliseconds; flip angle, 150 degrees; slice thickness, 2 mm; FOV, 240 240 mm2; matrix size, 256 160); diffusion tensor imaging (TR, 7600 milliseconds; TE, 91 milliseconds; slice thickness, 3 mm; slice space, 0 mm; FOV, 230 230 mm2; matrix size, 128 128; voxel size, 1.8 1.8 3 mm3; 20 directions); and blood oxygen level-dependent functional MRI (TR, 3000 milliseconds; TE, 30 milliseconds; FOV, 240 240 mm2; matrix size, 96 96; slice thickness, 3 mm) for the preoperative localization of language areas. The arcuate fasciculus and activations of the lan-
2.4. Intraoperative language mapping A constant-current generator (Epoch XP, Axon Systems Inc., Hauppauge, New York, USA) was used for intraoperative electrophysiologic monitoring during the procedure. MRI-safe subdermal needle electrodes (Chinese patent No. ZL201110074011.0) were inserted into the scalp and the target muscles to record compound muscle action potentials before craniotomy. For intraoperative direct cortical stimulation (DCS), a monophasic square-wave pulse was delivered at 60 Hz through a 5 mm wide bipolar electrode. The stimulation current ranged from 2 mA to 6 mA. While DCS was in progress, a 4- or 6-contact-strip subdural electrode was used to record after-discharge activity. The presence of after-discharge potentials indicated that the stimulation current was too high and the threshold would be decreased by 0.5–1 mA. This threshold was adopted as the upper limit of the sequential stimulation current. Using this ideal stimulation current, the patient completed three language tasks: counting from one to 50, picture naming, and word reading. Language deficits were classified as speech arrest, anomia, or alexia.19 Speech arrest was distinguished from dysarthria, which is caused by an involuntary muscle (mouth or pharyngeal muscle) contraction. Each cortical site was checked three times (Supplementary Fig. 1). If necessary, DCS and direct subcortical electrical stimulation were used for motor mapping. All of the positive sites were marked on the surface of the cortex with sterile tags, and the locations of these tags were recorded with photographs and navigational snapshots. Any adverse effects during the operation caused by electrical stimulation were recorded.
1282
J. Lu et al. / Journal of Clinical Neuroscience 20 (2013) 1280–1287
Fig. 1. A 48-year-old woman with a left insular astrocytoma (World Health Organization Grade II). (A, B) Task-based functional MRI with the co-registered 3-Dimensional T1 contrast data set before surgery showing the activation areas during (A) picture naming and (B) verb generation. The locations of language production between the tasks were the same in this case, and both were located superior to the tumor. (C) 3-Dimensional reconstruction of the arcuate fasciculus showing relationships with the tumor. (This figure is available in colour at www.sciencedirect.com.)
2.5. Intraoperative MRI procedure The purpose of iMRI in all cases was to evaluate the EOR, rather than just to update the neuronavigation data. Accordingly, the resection continued until the following occurred: (i) the surgeons deemed that the surgical goal was met, primarily through navigation and microscopy; (ii) the distance from the resection margin to the positive language cortical sites was less than 10 mm; or (iii) cortical motor mapping or subcortical motor pathway mapping was positive on the surgical margin. The iMRI was then prepared for scanning. The iMRI-integrated neurosurgical suite includes an operating room and a diagnostic room (Supplementary Fig. 2). The core equipment is a ceiling-mounted, movable 3.0 T magnet with a 70 cm working aperture. Prior to the iMRI device being moved into the operating room, the diagnostic room was closed for 30 minutes of laminar flow for sterilization. All of the instruments and tables, except for the MRI-safe subcutaneous needle electrodes (Supplementary Fig. 3A), were withdrawn by the nurses beyond the 5Gauss boundary line. To facilitate airway management and to make the awake craniotomy much less challenging in the movable MRI environment, we introduced the Mayo Clinic’s minimal draping25 technique to the IMRIS system (Supplementary Fig. 3B–D). After the MRI safe checklist was completed, the iMRI device was introduced into the operating room, and an interdissection MRI was obtained (Supplementary Fig. 3E and F). The sequences of structural MRI were the same as the preoperative ones. Once scanning was finished, the drape and the adhesive clear plastic drape were re-
moved, and the wound was redraped. If a residual tumor that was amenable to further resection was identified (Fig. 2), the navigational images were updated for further resection guidance. Because MRI-safe subcutaneous needle electrodes were used, brain mapping could be conducted again. Intraoperative scanning was repeated until the iMRI confirmed that maximal tumor resection had been achieved. 2.6. Postoperative language evaluation The same neuropsychologist performed postoperative language assessments before patient discharge (within 7–10 days) and at the 1-month and 3-month follow-up to demonstrate whether surgery resulted in improvement (the increase of AQ > 10 points), deterioration (the decrease of AQ > 10 points), or no change in language functions (increase or decrease 6 10 points) compared with the preoperative baseline evaluation. Patients who had postoperative impairment that had not recovered at 3 months were tested again at 6 months. 2.7. Control of volumetric resection and statistical analysis For contrast-enhancing gliomas, 3D MPRAGE images with gadolinium were used for the volume determination before and after resection. FLAIR imaging was used for nonenhancing gliomas. A postoperative MRI was obtained within 72 hours of the craniotomy procedure in those patients who had further resections but no iMRI confirmation. If there was no further resection, the final iMRI scan
Fig. 2. A 30-year-old man with a left frontal insular astrocytoma (World Health Organization Grade II). Axial fluid-attenuated inversion recovery (FLAIR) images showing tumor volumes. The preoperative (Pre-op) image shows the initial tumor volume to be 87.9 cm3. Intraoperative (Intra-op) image detected a residual tumor with a volume of 21.9 cm3, and the postoperative (Post-op) image obtained after additional resection confirmed a small remnant (6.0 cm3) at the posterior resection border. At the 1-month follow up, the remnant was stable. However, at the 3-month follow up, the residual tumor had disappeared after radiotherapy, and the patient exhibited no language deficits.
J. Lu et al. / Journal of Clinical Neuroscience 20 (2013) 1280–1287
1283
Fig. 3. A 40-year-old man with a left parietal glioblastoma multiforme (World Health Organization Grade IV). (A–D) Transaxial T1-weighted contrast-enhanced images. (A) Preoperative image showing the initial tumor (34.9 cm3). (B) Intraoperative image showing no residual tumor (final tumor volume: 0 cm3). The 1-month (C) and 3-month (D) follow-up images showing no tumor progression. No transient or permanent language deficits occurred. Photographs showing the intraoperative view (E) before and (F) after the resection performed awake under local anesthesia. (G) Schematic drawing showing the relationships between stimulated positive sites and the tumor. The eloquent cortical sites are marked by numbers or letters as follows: L: language cortex inducing speech arrest during counting, F: primary motor cortex of the face, H: primary motor cortex of the hand. (This figure is available in colour at www.sciencedirect.com.)
was considered the postoperative MRI (Fig. 3). In this cohort, contrast enhancement was shown in all high-grade gliomas (HGG) and was not shown in any low-grade gliomas (LGG). All pre- and postoperative tumoral segmentations were manually performed using the OsiriX (Pixmeo, Bernex, Switzerland) software tool by a neurosurgeon and were verified by an additional neurosurgeon.30 The EOR was calculated as follows: (preoperative tumor volume - postoperative tumor volume)/preoperative tumor volume.5 The patients were divided into two groups: patients without further tumor resection after iMRI (group 1) and patients who underwent the removal of additional tumor tissue after iMRI (group 2). For each group, we subdivided the members into the LGG (World Health Organization [WHO] Grade II) and HGG (WHO Grades III and IV) subgroups.31 In patients who underwent further tumor tissue removal (group 2), the final EOR was compared with the EOR obtained from the first intraoperative scan by the Wilcoxon signed-rank test, and a significance level of 0.05 was assumed (p < 0.05). Statistical analyses were performed using STATA software (version 10, StataCorp LP, College Station, Texas, USA). In this study, gross total resection (GTR) was defined as 100% resection of the tumor volume. 3. Results From March 2011 to November 2011, 39 patients were assessed for eligibility. Of these, two patients did not meet the inclusion criteria. Six patients were excluded because of intracranial hypertension (n = 5) and severe language deficits (n = 1). One patient refused to participate in the study. In total, 30 consecutive patients (21 men and nine women), with a median age of 45.5 years (range, 19–67 years), were prospectively evaluated (Table 1). The most common histopathologic diagnosis was diffuse astrocytoma (n = 13, 43.3%), followed by glioblastoma multiforme (n = 10, 33.3%), oligodendroglioma (n = 4, 13.3%), oligoastrocytoma (n = 2,
6.7%) and anaplastic oligodendroglioma (n = 1, 3.3%). Newly diagnosed gliomas were treated in 25 procedures, and recurrent gliomas were treated in five. The majority of the tumors were located in the left frontal lobe (n = 14, 46.7%), followed by the insular lobe (n = 8, 26.7%), the parietal lobe (n = 4, 13.3%) and the temporal lobe (n = 4, 13.3%). 3.1. Extent of resection In this study, all of the surgical procedures were performed by consultant neurosurgeons (J. Wu and Y. Mao). The first-look iMRI confirmed GTR in 11 (36.7%) patients. Of the 19 patients in which complete resection was not achieved as determined by the firstlook iMRI, further resection was not attempted in eight patients because of tumor involvement in the eloquent brain areas. Overall, GTR was achieved in 18 (60.0%) of the 30 patients. Thus, the use of iMRI increased the GTR rate from 36.7% to 60.0%, which represents a 23.3% increase. When the EOR was stratified according to tumor grade, GTR was achieved in 47.4% of the LGG and 81.8% of the HGG. Tables 2 and 3 list the volumetric assessment in the two groups, patients without further resection (group 1) and patients with further resection (group 2). For the 12 patients with LGG in group 1 (Table 2), the median EOR was 96.3% (range, 73.9–100%), and GTR was accomplished in five patients. Six of seven patients with HGG achieved GTR. Among the 19 patients in group 1, the median pre- and postoperative tumor volumes were 62.3 cm3 (range, 8.4–216.7 cm3) and 0 cm3 (range, 0–56.5 cm3), respectively. The median EOR for all 19 patients was 100% (range, 73.9–100%). The gliomas in group 2 (Table 3) included seven LGG and four HGG that underwent further resection. Because iMRI was used, GTR was accomplished in seven of 11 patients. Remnants were present in four patients (three LGG and one HGG) after additional tumor resection as a result of the invasion of critical structures. In
1284
J. Lu et al. / Journal of Clinical Neuroscience 20 (2013) 1280–1287 Table 1 Demographics of patient undergoing tumor resection with awake language maping and iMRI Variable
Value
Age (years) Median Range
45.5 19–67
Male sex: No. (%)
21 (70.0)
Signs and symptoms at presentation: No. (%) Seizure Headache Language deficit Paresthesia
16 (53.3) 9 (30.0) 7 (23.3) 1 (3.3)
WHO tumor grade and histologic type: No. (%) II, diffuse astrocytoma II, oligodendroglioma II, oligoastrocytoma III, anaplastic oligodendroglioma IV, glioblastoma multiforme
13 (43.3) 4 (13.3) 2 (6.7) 1 (3.3) 10 (33.3)
Tumor location: No. (%) Frontal Parietal Temporal Insular
14 (46.7) 4 (13.3) 4 (13.3) 8 (26.7)
Tumor volume (cm3) Median Range
60.0 8.4–216.7
Gross total resection according to WHO tumor grade31 (in the first iMRI): No./total No. (%) All grades I or II III or IV
11/30 (36.7) 5/19 (26.3) 6/11 (54.5)
Gross total resection according to WHO tumor grade: No./total No. (%) All grades I or II III or IV
18/30 (60.0) 9/19 (47.4) 9/11 (81.8)
New or increased postoperative language deficit: No./total No. (%) 1 week 1 month 3 months 6 months
12 (40.0) 5 (16.7) 2 (6.7) 1 (3.3)
iMRI = intraoperative MRI, WHO = World Health Organization.
Table 2 Summary of volumetric assessment in 19 patients without further tumor resection after iMRI Tumor grade
No. of patients
Median initial tumor volume cm3 (range)
Median residue volume on iMRI cm3 (range)
Median EOR (range)
No. of patients with GTR
LGG HGG Total
12 7 19
68.0 (9.1–216.7) 57.0 (8.4–81.9) 62.3 (8.4–216.7)
2.7 (0–56.5) 0 (0–3.7) 0 (0–56.5)
96.3% (73.9–100%) 100% (95.5–100%) 100% (73.9–100%)
5 6 11
EOR = extent of resection, GTR = gross total resection, HGG = high-grade glioma, iMRI = intraoperative MRI, LGG = low-grade glioma.
the LGG subgroup, the final EOR were significantly higher compared with those obtained from the first iMRI (p = 0.02). There was no significant difference between the EOR determined from the first iMRI and the final EOR (p = 0.07) in the HGG subgroup, although the median EOR increased from 95.8% (range, 82.8– 96.3%) to 100% (range, 94.2–100%). Among the 11 patients, the median EOR increased from 92.5% (range, 75.1–97.0%) to 100% (range, 92.6–100%), which was statistically significant (p < 0.01).
follow up, five (16.7%) patients had not returned to their baseline function. By 3 months, only two (6.7%) of the 30 patients had new language deficits (anomia). At the 6-month follow up, only one (3.3%) had permanent language deficits (Table 4).
3.2. Functional outcome
Although four patients presented with focal seizures (myoclonic twitches) during mapping, cold saline was able to stop the attacks. There were no generalized seizures during any of the procedures. There were no complications related to the iMRI or any adverse events caused by the high-field MRI. All of the patients tolerated the processes well and without complaints of intraoperative moderate or severe pain.32
During the preoperative language assessment, seven patients had language deficits that were not significant enough to prevent them from performing intraoperative tasks. One week after surgery, 12 out of 30 (40.0%) patients (Table 4) demonstrated deterioration (n = 4) or new language deficits (n = 8). At the 1-month
3.3. Complications
1285
J. Lu et al. / Journal of Clinical Neuroscience 20 (2013) 1280–1287 Table 3 Summary of volumetric assessment in 11 patients who underwent removal of additional tumor tissue after intraoperative MRI Tumor grade
No. of patients
Median initial tumor volume cm3 (range)
Median residue volume in the first iMRI cm3 (range)
Median EOR in the first iMRI (range)
Median final tumor volume cm3 (range)
Median final EOR (range)
pvalue
No. of patients with GTR (%)
LGG HGG Total
7 4 11
57.7 (35.8–103.9) 59.8 (29.1–80.9) 57.7 (29.1–103.9)
5.7 (2.1–21.9) 3.0 (2.2–5.0) 3.0 (1.4–21.9)
89.0% (75.1–97.0%) 95.8% (82.8–96.3%) 92.5% (75.1–97.0%)
0 (0–7.7) 0 (0–1.7) 0 (0–7.7)
100% (92.6–100%) 100% (94.2–100%) 100% (92.6–100%)
0.02 0.07 <0.01
4 3 7
EOR = extent of resection, GTR = gross total resection, HGG = high-grade glioma, iMRI = intraoperative MRI, LGG = low-grade glioma.
Table 4 Clinical characteristics and results of 12 patients who suffered from transient deficits following resection for gliomas in language areas Patient No.
Age (years), Sex
Location
6 8 9 10 12 13 14 15 27 28 29 30
66, 49, 41, 47, 40, 48, 27, 63, 65, 58, 46, 36,
T P I P I I I I Fr I T I
M M M M F F M M M M M M
Pathology (WHO grade)31
Initial tumor volume (cm3)
EOR on the first iMRI (%)
Final EOR (%)
Pre-op AQ
Post-op AQ 1 week
1 month
3 months
GBM (IV) GBM (IV) Astrocytoma (II) Anaplastic Oligodendroglioma (III) Astrocytoma (II) Astrocytoma (II) Astrocytoma (II) Astrocytoma (II) GBM(IV) Oligodendroglioma (II) Oligoastrocytoma(II) Astrocytoma (II)
80.9 29.1 98.6 64.7 103.9 48.7 123.9 73.7 50.9 62.3 47.9 96.4
96.3 82.8 89.0 100.0 80.1 88.3 78.3 87.9 95.7 96.5 97.1 93.9
100.0 94.2 98.3 100.0 92.6 100.0 78.3 87.9 100.0 96.5 100.0 93.9
83.2 78.8 96.6 73.9 97.4 97.5 99.4 98.4 88.4 99.0 95.8 98.5
50.9 46.6 46.3 50.6 27.4 35.1 33.9 49.5 66.1 33.0 73.3 77.5
76.2 73.1 93.0 72.0 73.3 90.2 97.6 69.0 76.7 79.6 98.8 84.7
82.3 81.0 91.5 82.5 84.8 88.6 96.6 66.6 93.5 96.6 97.0 93.8
6 months
87.8
54.5
AQ = aphasia quotient, EOR = extent of resection, F = female, Fr = frontal, GBM = glioblastoma multiforme, I = insular, iMRI = intraoperative magnetic resonance imaging, M = male, P = parietal, Post-op = postoperative, Pre-op = preoperative, T = temporal, WHO = World Health Organization.
Table 5 Comparison with previous studies using both high-field intraoperative MRI guidance and awake craniotomy Authors, year of publication
No. of cases
Strength of iMRI
iMRI concepts
iMRI draping protocols
Airway management during iMRI scan
iMRI-safe subcutaneous needle electrodes
Volumetric analysis
Nabavi et al., 200924 Goebel et al., 201023 Weingarten et al., 200926 Parny et al., 201025 Leuthardt et al., 201127 Lu et al. (present study), 2013
38
1.5 T
Fixed magnet
Uncovering the patient’s face (no details)
ND
ND
No
10
1.5 T
ND
No
1.5 T
Trimming drapes hanging below the operating table (no details) Minimal draping
ND
1
No LMA
ND
No
12
1.5 T
ND
No
3.0 T
Standard draping protocol (the head and upper body are wrapped in sterile drapes) Minimal draping
LMA
30
Fixed magnet Fixed magnet Movable magnet Movable magnet
No LMA
Yes
Yes
25
iMRI = intraoperative magnetic resonance imaging, LMA = laryngeal mask airway, ND = no data.
4. Discussion Despite a lack of class I evidence, an increasing number of studies suggest that more extensive surgical resection is associated with longer life expectancy for both LGG and HGG.7,33 However, extensive resection carries the risk of neurologic morbidity, which affects the quality of life and subsequent overall survival.11 In addition, quality of life is currently deemed to be the highest priority in brain tumor management. Thus, the precise localization and preservation of eloquent areas during surgery is vital, particularly the conservation of language functionality, which is one of the most fundamental skills required for social interaction. 4.1. Localization of language areas Several methods have been applied to the localization of language areas. Language functional magnetic resonance imaging
(fMRI) and brain mapping using DCS are the most common techniques to localize language areas. Although fMRI is safe, noninvasive and repeatable, its reliability for localization remains uncertain.34 Previous studies have shown great variability in the sensitivity (59–100%) and specificity (0–97%) of fMRI compared with DCS34–37 and have confirmed via different language tasks that the distribution of language activation is widespread. As the gold standard, intraoperative language mapping by DCS has been shown to be a good predictor of functional recovery in Indo-European languages such as English and French.14,19–21 Nonetheless, Chinese differs notably from alphabetic languages. Whereas alphabetic words have a linear structure made up of letters, Chinese words have a square, nonlinear configuration that uses pictographic characters as basic units.38 Tan et al.39,40 reported that the process of Chinese involves more brain areas than that of English, so they suspected that injury to or surgery of the brain in Chinese-speakers may result in a higher rate of language deficits. In
1286
J. Lu et al. / Journal of Clinical Neuroscience 20 (2013) 1280–1287
this present study, we assessed the functional outcome of glioma surgery on language areas in Chinese speakers using this gold standard. Unfortunately, we had a higher incidence of transient postoperative language deficits (40.0% at 1 week) compared with two of the largest studies describing awake language mapping in English speakers (Sanai et al., 22.4%19 and Kim et al., 36%20). However, the long-term functional outcomes were similar to the above studies (16.7% at 1 month and 3.3% at 6 months for this study). We suspect that this higher incidence in transient deficits can be attributed to differences between the Chinese and English languages. Of course, we cannot draw a firm conclusion because of the small number of patients enrolled. With regard to EOR, in this study, GTR was achieved in 60% of the patients, which did not demonstrate an obvious advantage for iMRI. If P95%20 was used to calculate the GTR in all cases, 73.3% of the patients achieved complete resection, which is higher than in the two previous studies.19,20 If the assumed 90% EOR threshold for LGG were applied,5 the surgical survival advantage improves from 12 (63.2%) to 16 (84.2%) patients in 19 LGG after iMRI. We believe that this difference can be attributed to the contribution of iMRI. 4.2. Contribution of iMRI and volumetric analysis Over the past decade, a growing body of literature has focused on the added value of iMRI. Approximately 12 non-randomized cohort studies have demonstrated that iMRI can increase the EOR for glioblastoma multiforme, which has also been well reviewed by Kubben et al.4 The evidence-based grade for recommendation of the clinical efficiency of iMRI currently stands at level B.4,6 There were no randomized, controlled trials that provided evidence for the implementation of iMRI in glioma surgery until 2011, when Senft et al.41 conducted the first randomized trial investigating the application of iMRI in patients with contrast-enhancing gliomas. They confirmed the beneficial effects of 0.15 T iMRI for GTR compared with conventional microsurgery. The rate of GTR by volumetric analysis in the iMRI group was obviously higher than that in the control group (i.e., 96% versus 68%, p = 0.02). In addition, several non-randomized cohort studies have also investigated the contribution of iMRI to EOR using a volumetric assessment.42–44 Kuhnt et al.44 reported what was previously the largest dataset, which demonstrated that the percentage of complete resections increased from 31.7% to 38.6%. Our data show that the percentage of GTR increased from 36.7 to 60.0%. The final GTR in this study was higher, which was perhaps attributable to the higher number of HGG, which had a higher rate of GTR (81.8%). Furthermore, the majority of the published reports, including volumetric assessments, have also reported rates of continued surgery, ranging roughly from 26% to 53%. Our rate of 36.7% falls within this range. Interestingly, it seems that a higher rate of continued surgery is associated with a higher number of patients achieving GTR. The aforementioned studies included patients with gliomas involving non-eloquent areas, therefore the functional outcomes of our study are not directly comparable.
combination were confirmed by Weingarten et al.26 and minimal draping was reported in the Mayo Clinic’s case report (n = 1).25 There were several features in common in the above-mentioned studies. The initial experiences were all conducted in a 1.5 T MRI environment, and all physicians preferred transporting the patients over moving the MR scanner itself. However, in a recent case study, Leuthardt et al.27 described the integration of awake craniotomy in a movable high-field (1.5 T) MRI environment. The differences between previous studies and the present study are shown in Table 5. The highlighted point is the airway management and the resulting pattern of draping. As Leuthardt et al.27 described, airway management is critical to this combination. The awake anesthesia approach that they adopted was similar to the ‘‘asleepawake-asleep’’ technique,45 which places a laryngeal mask over the airway during scanning. Although this technique can provide a secure airway and hyperventilate the patient during the operation, it involves elaborate intraoperative airway maneuvers and requires a particular head positioning. Furthermore, topical anesthesia during re-intubation of the upper airway may result in the loss of airway reflexes. The MAC46 technique that we adopted during the procedure was comparatively easy to perform and allowed the patient to be awakened at any time. This protocol provided adequate sedation and analgesia without a laryngeal mask airway. Nevertheless, adopting MAC in high-field iMRI units is problematic for standard draping protocols because of airway protection.25,27 The minimal draping technique25 effectively solved these issues and was confirmed as safe, sterile and simple. Hence, we introduced this technique into the movable 3.0 T iMRI environment and validated that it could be safely and effectively applied. 4.4. Study limitations There were some limitations to the current study. First, the small number of patients limited our ability to draw any conclusion on whether the combined procedure affected the eventual outcomes. However, our data could contribute to a growing worldwide database, and larger datasets are being collected. Furthermore, we did not perform language subcortical stimulation in our series because it is time consuming. Neuronavigation of the arcuate fasciculus was applied in the requisite cases. Finally, this study reported a case series, and we did not include comparisons with a control group. To unequivocally demonstrate the value of iMRI, a prospective, randomized, controlled clinical trial is underway at our center.
5. Conclusion In this study, we present our clinical experience with 30 patients who underwent glioma resection aided by the combination of awake language mapping and iMRI guidance. It demonstrates the potential utility of combining awake craniotomy with iMRI. It is safe and reliable to perform awake surgery in a movable iMRI.
4.3. Comparison with previous studies combining iMRI guidance and awake craniotomy
Acknowledgments
Although these two techniques have been widely used, only four neurosurgical units have reported their preliminary clinical experiences with the combined use of awake craniotomy and brain mapping in high-field iMRI surgical suites (Table 5). The first integration of iMRI and awake craniotomy was undertaken by Nabavi et al.24 in 2009. Thereafter, this group also prospectively evaluated 25 patients for depression and anxiety and confirmed that iMRI appears to have no additional significant impact on subjective patient perceptions.23 At the same time, the safety and reliability of this
The authors would like to thank the nurses (Yanwei Gong, Chunmei Li, Ye Wang) and iMRI technician (Zhong Yang) of our team for their perfect cooperation during the complex procedures; Jianbing Shi for technical support of neuronavigation; Geng Xu for intraoperative neurophysiologic support; Qihao Guo and Yan Zhou for the language assessment; Yanyan Song for the statistical analysis and Ian F. Parney for the advice on minimal draping. This work was supported by the Ministry of Health of China (2010–2012), the National Natural Science Foundation of China (Project No.
J. Lu et al. / Journal of Clinical Neuroscience 20 (2013) 1280–1287
81071117) and (XBR2011022).
the
Shanghai
Municipal
Health
Bureau
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jocn.2012.10.042. References 1. Wen PY, Kesari S. Malignant gliomas in adults. N Engl J Med 2008;359:492–507. 2. Claus EB, Horlacher A, Hsu L, et al. Survival rates in patients with low-grade glioma after intraoperative magnetic resonance image guidance. Cancer 2005;103:1227–33. 3. Claus EB, Black PM. Survival rates and patterns of care for patients diagnosed with supratentorial low-grade gliomas: data from the SEER program, 1973– 2001. Cancer 2006;106:1358–63. 4. Kubben PL, ter Meulen KJ, Schijns OE, et al. Intraoperative MRI-guided resection of glioblastoma multiforme: a systematic review. Lancet Oncol 2011;12:1062–70. 5. Smith JS, Chang EF, Lamborn KR, et al. Role of extent of resection in the longterm outcome of low-grade hemispheric gliomas. J Clin Oncol 2008;26:1338–45. 6. Kuhnt D, Becker A, Ganslandt O, et al. Correlation of the extent of tumor volume resection and patient survival in surgery of glioblastoma multiforme with highfield intraoperative MRI guidance. Neuro Oncol 2011;13:1339–48. 7. Sanai N, Berger MS. Glioma extent of resection and its impact on patient outcome. Neurosurgery 2008;62:753–64. 8. Diez Valle R, Tejada Solis S, Idoate Gastearena MA, et al. Surgery guided by 5aminolevulinic fluorescence in glioblastoma volumetric analysis of extent of resection in single-center experience. J Neurooncol 2011;102:105–13. 9. Lacroix M, Abi-Said D, Fourney DR, et al. A multivariate analysis of 416 patients with glioblastoma multiforme: prognosis, extent of resection, and survival. J Neurosurg 2001;95:190–8. 10. Vecht CJ, Avezaat CJ, van Putten WL, et al. The influence of the extent of surgery on the neurological function and survival in malignant glioma. A retrospective analysis in 243 patients. J Neurol Neurosurg Psychiatry 1990;53:466–71. 11. McGirt MJ, Mukherjee D, Chaichana KL, et al. Association of surgically acquired motor and language deficits on overall survival after resection of glioblastoma multiforme. Neurosurgery 2009;65:463–9 discussion 469–470. 12. Lang MJ, Kelly JJ, Sutherland GR. A moveable 3-Tesla intraoperative magnetic resonance imaging system. Neurosurgery 2011;68(1 Suppl Operative):168–79. 13. Nimsky C, Ganslandt O, Von Keller B, et al. Intraoperative high-field-strength MR imaging: implementation and experience in 200 patients. Radiology 2004;233:67–78. 14. Ojemann G, Ojemann J, Lettich E, et al. Cortical language localization in left, dominant hemisphere. An electrical stimulation mapping investigation in 117 patients. J Neurosurg 1989;71:316–26. 15. Duffau H, Capelle L, Sichez N, et al. Intraoperative mapping of the subcortical language pathways using direct stimulations. An anatomo-functional study. Brain 2002;125:199–214. 16. Zhao Y, Chen X, Wang F, et al. Integration of diffusion tensor-based arcuate fasciculus fibre navigation and intraoperative MRI into glioma surgery. J Clin Neurosci 2012;19:255–61. 17. Sun GC, Chen XL, Zhao Y, et al. Intraoperative MRI with integrated functional neuronavigation-guided resection of supratentorial cavernous malformations in eloquent brain areas. J Clin Neurosci 2011;18:1350–4. 18. Nimsky C, Ganslandt O, Hastreiter P, et al. Intraoperative compensation for brain shift. Surg Neurol 2001;56(6):357–64 discussion 364–355. 19. Sanai N, Mirzadeh Z, Berger MS. Functional outcome after language mapping for glioma resection. N Engl J Med 2008;358:18–27. 20. Kim SS, McCutcheon IE, Suki D, et al. Awake craniotomy for brain tumors near eloquent cortex: correlation of intraoperative cortical mapping with neurological outcomes in 309 consecutive patients. Neurosurgery 2009;64:836–45 discussion 345–836. 21. Duffau H, Capelle L, Sichez J, et al. Intra-operative direct electrical stimulations of the central nervous system: the Salpetriere experience with 60 patients. Acta Neurochir (Wien) 1999;141:1157–67.
1287
22. Lu JF, Zhang J, Wu JS, et al. Awake craniotomy and intraoperative language cortical mapping for eloquent cerebral glioma resection: preliminary clinical practice in 3.0 T intraoperative magnetic resonance imaging integrated surgical suite. Zhonghua Wai Ke Za Zhi 2011;49:693–8. 23. Goebel S, Nabavi A, Schubert S, et al. Patient perception of combined awake brain tumor surgery and intraoperative 1.5-T magnetic resonance imaging: the Kiel experience. Neurosurgery 2010;67:594–600 discussion 600. 24. Nabavi A, Goebel S, Doerner L, et al. Awake craniotomy and intraoperative magnetic resonance imaging: patient selection, preparation, and technique. Top Magn Reson Imaging 2009;19:191–6. 25. Parney IF, Goerss SJ, McGee K, et al. Awake craniotomy, electrophysiologic mapping, and tumor resection with high-field intraoperative MRI. World Neurosurg 2010;73:547–51. 26. Weingarten DM, Asthagiri AR, Butman JA, et al. Cortical mapping and frameless stereotactic navigation in the high-field intraoperative magnetic resonance imaging suite technical note. J Neurosurg 2009;111:1185–90. 27. Leuthardt EC, Lim CC, Shah MN, et al. Use of movable high-field-strength intraoperative magnetic resonance imaging with awake craniotomies for resection of gliomas: preliminary experience. Neurosurgery. 2011;69:194–205 discussion 205–196. 28. Gao SR, Chu YF, Shi SQ, et al. A standardization research of the aphasia battery of Chinese. Chinese Mental Health Journal 1992;6:125–8. 29. Kertesz A. The Western Aphasia Battery test manual: New York: Grune & Stratton; 1982. 30. Rosset A, Spadola L, Ratib O. OsiriX: an open-source software for navigating in multidimensional DICOM images. J Digit Imaging 2004;17:205–16. 31. Louis DN, Ohgaki H, Wiestler OD, et al. The 2007 WHO classification of tumours of the central nervous system. Acta Neuropathol 2007;114:97–109. 32. Hartrick CT, Kovan JP, Shapiro S. The numeric rating scale for clinical pain measurement: a ratio measure? Pain Pract 2003;3:310–6. 33. Sanai N, Berger MS. Operative techniques for gliomas and the value of extent of resection. Neurotherapeutics 2009;6:478–86. 34. Giussani C, Roux FE, Ojemann J, et al. Is preoperative functional magnetic resonance imaging reliable for language areas mapping in brain tumor surgery? Review of language functional magnetic resonance imaging and direct cortical stimulation correlation studies. Neurosurgery 2010;66:113–20. 35. Bizzi A, Blasi V, Falini A, et al. Presurgical functional MR imaging of language and motor functions: validation with intraoperative electrocortical mapping. Radiology 2008;248:579–89. 36. Rutten GJ, Ramsey NF, van Rijen PC, et al. Development of a functional magnetic resonance imaging protocol for intraoperative localization of critical temporoparietal language areas. Ann Neurol 2002;51:350–60. 37. Spena G, Nava A, Cassini F, et al. Preoperative and intraoperative brain mapping for the resection of eloquent-area tumors. A prospective analysis of methodology, correlation, and usefulness based on clinical outcomes. Acta Neurochir 2010;152:1835–46. 38. Siok WT, Perfetti CA, Jin Z, et al. Biological abnormality of impaired reading is constrained by culture. Nature 2004;431:71–6. 39. Tan LH, Laird AR, Li K, et al. Neuroanatomical correlates of phonological processing of Chinese characters and alphabetic words: a meta-analysis. Hum Brain Mapp 2005;25:83–91. 40. Tan LH, Spinks JA, Gao JH, et al. Brain activation in the processing of Chinese characters and words: a functional MRI study. Hum Brain Mapp 2000;10:16–27. 41. Senft C, Bink A, Franz K, et al. Intraoperative MRI guidance and extent of resection in glioma surgery: a randomised, controlled trial. Lancet Oncol 2011;12:997–1003. 42. Hatiboglu MA, Weinberg JS, Suki D, et al. Impact of intraoperative high-field magnetic resonance imaging guidance on glioma surgery: a prospective volumetric analysis. Neurosurgery. 2009;64:1073–81 discussion 1081. 43. Schneider JP, Trantakis C, Rubach M, et al. Intraoperative MRI to guide the resection of primary supratentorial glioblastoma multiforme–a quantitative radiological analysis. Neuroradiology 2005;47:489–500. 44. Kuhnt D, Ganslandt O, Schlaffer SM, et al. Quantification of glioma removal by intraoperative high-field magnetic resonance imaging – an update. Neurosurgery 2011;69:852–62 discussion 862–863. 45. Huncke K, Van de Wiele B, Fried I, et al. The asleep-awake-asleep anesthetic technique for intraoperative language mapping. Neurosurgery. 1998;42:1312–6 discussion 1316–1317. 46. Berkenstadt H, Perel A, Hadani M, et al. Monitored anesthesia care using remifentanil and propofol for awake craniotomy. J Neurosurg Anesthesiol 2001;13:246–9.
Contents lists available at SciVerse ScienceDirect
Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn
Clinical Study
Awake language mapping and 3-Tesla intraoperative MRI-guided volumetric resection for gliomas in language areas Junfeng Lu a, Jinsong Wu a,⇑, Chengjun Yao a, Dongxiao Zhuang a, Tianming Qiu a, Xiaobing Hu b, Jie Zhang a, Xiu Gong a, Weimin Liang b, Ying Mao a, Liangfu Zhou a a b
Division of Glioma Surgery, Department of Neurosurgery, Huashan Hospital, Shanghai Medical College, Fudan University, Shanghai 200040, China Department of Anesthesiology, Huashan Hospital, Shanghai Medical College, Fudan University, Shanghai, China
a r t i c l e
i n f o
Article history: Received 14 August 2012 Accepted 7 October 2012
Keywords: Awake surgery Direct electrical stimulation Glioma Intraoperative MRI Volumetric analysis
a b s t r a c t The use of both awake surgery and intraoperative MRI (iMRI) has been reported to optimize the maximal safe resection of gliomas. However, there has been little research into combining these two demanding procedures. We report our unique experience with, and methodology of, awake surgery in a movable iMRI system, and we quantitatively evaluate the contribution of the combination on the extent of resection (EOR) and functional outcome of patients with gliomas involving language areas. From March 2011 to November 2011, 30 consecutive patients who underwent awake surgery with iMRI guidance were prospectively investigated. The EOR was assessed by volumetric analysis. Language assessment was conducted before surgery and 1 week, 1 month, 3 months and 6 months after surgery using the Aphasia Battery of Chinese. Awake language mapping integrated with 3.0 Tesla iMRI was safely performed for all patients. An additional resection was conducted in 11 of 30 patients (36.7%) after iMRI. The median EOR significantly increased from 92.5% (range, 75.1–97.0%) to 100% (range, 92.6–100%) as a result of iMRI (p < 0.01). Gross total resection was achieved in 18 patients (60.0%), and in seven of those patients (23.3%), the gross total resection could be attributed to iMRI. A total of 12 patients (40.0%) suffered from transient language deficits; however, only one (3.3%) patient developed a permanent deficit. This study demonstrates the potential utility of combining awake craniotomy with iMRI; it is safe and reliable to perform awake surgery using a movable iMRI. Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction Gliomas are the most common primary brain tumors and are associated with a poor prognosis.1–3 The standard treatment for gliomas is tumor resection followed by adjuvant radiotherapy and chemotherapy. An increasing body of literature4–7 demonstrates that the extent of resection (EOR) is a predictor of survival, despite a lack of class I evidence.8–10 However, the aggressive resection of gliomas located in eloquent areas carries the risk of neurologic deficits that can lead to a loss of quality of life, which may subsequently affect survival.11 Thus, for neurosurgeons, the main challenge in glioma surgery is achieving the maximal tumor resection while still preserving eloquent areas. Various advanced techniques and methods, including neuronavigation, intraoperative MRI (iMRI), 5-aminolevulinic acid and electrocortical mapping, have facilitated the maximal safe resection of gliomas in eloquent areas.12–17 Image guidance using neuronavigation, which is based on preoperative imaging, has been used extensively for several ⇑ Corresponding author. Tel./fax: +86 215 288 8771. E-mail address: [email protected] (J. Wu). 0967-5868/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jocn.2012.10.042
decades to aid in the resection of intracranial tumors. However, intraoperative brain shift poses a limitation for conventional neuronavigation. Therefore, iMRI has attracted increasing interest in the past decade because it effectively detects tumor remnants and compensates for intraoperative brain deformation.18 However, with the application of iMRI alone, it is difficult to preserve neurologic function reliably and to delineate a real-time safe boundary for tumor resection, particularly in gliomas involving language areas. Currently, awake language mapping is generally accepted as the gold standard for language localization and has been used with success in several languages of the Indo-European family.14,19–21 However, the feasibility and reliability of this technique as applied to the Chinese population (that is, Sino-Tibetan family speakers), have seldom been evaluated quantitatively.22 In addition, although both iMRI and awake language mapping are independent and established techniques, few studies23–27 have reported their combination. Experience in combining these two complicated modalities is limited, and the evidence is not sufficiently robust to draw a conclusion regarding the effectiveness of this combination.
J. Lu et al. / Journal of Clinical Neuroscience 20 (2013) 1280–1287
In this prospective study, we describe our approach and experience with the integration of awake language mapping and movable 3.0 Tesla (T) iMRI for the resection of gliomas in language areas. We aimed to both increase the EOR and to decrease the postoperative morbidity by combining these techniques. 2. Patients and methods 2.1. Patients Patients with tumors located in language areas as indicated by preoperative MRI were considered for tumor resection aided by the combination of iMRI and awake language mapping. Patients were selected using the following inclusion criteria: (i) the patients were adults of Han ethnicity; (ii) they usually spoke Chinese in daily life; (iii) they were right-hand dominant; and (iv) cerebral gliomas were preoperatively suspected. The exclusion criteria included the following: (i) patients with contraindications to MRI scanning, intraoperative neurophysiologic monitoring, or awake craniotomy (including patients with pacemakers, obstructive sleep apnea or severe intracranial hypertension); and (ii) patients whose pathologic diagnoses were not gliomas. All of the patients underwent language functional assessment using the Aphasia Battery of Chinese,28 which is the Chinese standardized adaptation of the Western Aphasia Battery. The Aphasia Quotient (AQ) score (spontaneous speech, comprehension, repetition and naming) was used to measure language ability. The preoperative cognitive state was also assessed using the Mini Mental State Examination.24 Patients with severe language deficits (AQ < 50) or cognitive disorders (Mini Mental State Examination score < 23) were excluded.24 Written informed consent was obtained from all patients. This study was undertaken at the Huashan Hospital (Shanghai, China) with approval from the Huashan Institutional Review Board.
1281
guage areas were then reconstructed into 3D objects and superimposed onto the structural data (Fig. 1A–C) using a post-processing workstation (Syngo MultiModality Workplace, Siemens AG). The 3D relationships between the lesion and the language areas were provided to the neurosurgeons for decision making. On the day of surgery, the reconstructed series were imported into the surgical navigation system.
2.3. Awake craniotomy A monitored anesthesia care (MAC) approach was adopted for all patients. After the anesthesiologists administered premedication by infusing 0.02–0.03 mg/kg midazolam and 5 mg tropisetron, they prepared the patient with intravenous lines, a central venous catheter, an arterial line, and a urethral catheter. The supraorbital, supratrochlear, zygomaticotemporal, auriculotemporal, greater occipital, and lesser occipital nerves of both sides were then blocked using a mixture of lidocaine (0.67%) and ropivacaine (0.5%). Once the patient was brought into moderate sedation with boluses of intravenous propofol, the head was fixed into its fitted position using a custom-designed high-field MRI-safe head holder (DORO Radiolucent Headrest System, Pro Med Instruments GmbH, Freiburg, Germany), which was integrated with an IMRIS operating room table and head coils (IMRIS). Once the scalp was prepared and draped, remifentanil (0.01–0.03 lg/Kg/min) or dexmedetomidine (0.1–0.7 lg/Kg/h) was administered for analgesia. To minimize brain swelling, mannitol (1 g/kg) was intravenously infused before the dura was opened. A low dose of remifentanil (0.01 lg/ Kg/min) or dexmedetomidine (0.1 lg/Kg/h) was administered during mapping. Because of the minimal draping, as described below, no laryngeal mask airway or endotracheal tubing was applied during the iMRI acquisition phase in these patients. Throughout the operation, supplemental inspired oxygen was delivered by nasal cannula or facemask.
2.2. Preoperative examination and evaluation Cognitive status and language function were preoperatively assessed by a neuropsychologist as previously mentioned. The patients were diagnosed with language deficits if they scored below 93.8 on the AQ preoperatively.29 Preoperative consultations by neurosurgeons and anesthesiologists occurred on the day before the operation. Patients were trained by neurosurgeons to perform the tasks of counting, naming objects and reading single words. Anesthesiologists were in charge of the interpretation for intraoperative discomfort and cooperation. Preoperative brain images were obtained in the diagnostic room of an iMRI-integrated neurosurgical suite (IMRIS, Winnipeg, Manitoba, Canada) using a movable 3.0 T scanner (MAGNETOM Verio 3.0 T, Siemens AG, Erlangen, Germany) 1 day prior to surgical intervention. The imaging protocol included a contrast-enhanced threedimensional (3D) magnetization-prepared rapid-gradient echo (MPRAGE) sequence (time of repetition (TR), 1900 milliseconds; time of echo (TE), 2.93 milliseconds; flip angle, 9 degrees; field of view (FOV), 250 250 mm2; matrix size, 256 215; slice thickness, 1 mm; acquisition averages, 1) or a fluid-attenuated inversion recovery (FLAIR) sequence (TR, 9000 milliseconds; TE, 96 milliseconds; TI, 2500 milliseconds; flip angle, 150 degrees; slice thickness, 2 mm; FOV, 240 240 mm2; matrix size, 256 160); diffusion tensor imaging (TR, 7600 milliseconds; TE, 91 milliseconds; slice thickness, 3 mm; slice space, 0 mm; FOV, 230 230 mm2; matrix size, 128 128; voxel size, 1.8 1.8 3 mm3; 20 directions); and blood oxygen level-dependent functional MRI (TR, 3000 milliseconds; TE, 30 milliseconds; FOV, 240 240 mm2; matrix size, 96 96; slice thickness, 3 mm) for the preoperative localization of language areas. The arcuate fasciculus and activations of the lan-
2.4. Intraoperative language mapping A constant-current generator (Epoch XP, Axon Systems Inc., Hauppauge, New York, USA) was used for intraoperative electrophysiologic monitoring during the procedure. MRI-safe subdermal needle electrodes (Chinese patent No. ZL201110074011.0) were inserted into the scalp and the target muscles to record compound muscle action potentials before craniotomy. For intraoperative direct cortical stimulation (DCS), a monophasic square-wave pulse was delivered at 60 Hz through a 5 mm wide bipolar electrode. The stimulation current ranged from 2 mA to 6 mA. While DCS was in progress, a 4- or 6-contact-strip subdural electrode was used to record after-discharge activity. The presence of after-discharge potentials indicated that the stimulation current was too high and the threshold would be decreased by 0.5–1 mA. This threshold was adopted as the upper limit of the sequential stimulation current. Using this ideal stimulation current, the patient completed three language tasks: counting from one to 50, picture naming, and word reading. Language deficits were classified as speech arrest, anomia, or alexia.19 Speech arrest was distinguished from dysarthria, which is caused by an involuntary muscle (mouth or pharyngeal muscle) contraction. Each cortical site was checked three times (Supplementary Fig. 1). If necessary, DCS and direct subcortical electrical stimulation were used for motor mapping. All of the positive sites were marked on the surface of the cortex with sterile tags, and the locations of these tags were recorded with photographs and navigational snapshots. Any adverse effects during the operation caused by electrical stimulation were recorded.
1282
J. Lu et al. / Journal of Clinical Neuroscience 20 (2013) 1280–1287
Fig. 1. A 48-year-old woman with a left insular astrocytoma (World Health Organization Grade II). (A, B) Task-based functional MRI with the co-registered 3-Dimensional T1 contrast data set before surgery showing the activation areas during (A) picture naming and (B) verb generation. The locations of language production between the tasks were the same in this case, and both were located superior to the tumor. (C) 3-Dimensional reconstruction of the arcuate fasciculus showing relationships with the tumor. (This figure is available in colour at www.sciencedirect.com.)
2.5. Intraoperative MRI procedure The purpose of iMRI in all cases was to evaluate the EOR, rather than just to update the neuronavigation data. Accordingly, the resection continued until the following occurred: (i) the surgeons deemed that the surgical goal was met, primarily through navigation and microscopy; (ii) the distance from the resection margin to the positive language cortical sites was less than 10 mm; or (iii) cortical motor mapping or subcortical motor pathway mapping was positive on the surgical margin. The iMRI was then prepared for scanning. The iMRI-integrated neurosurgical suite includes an operating room and a diagnostic room (Supplementary Fig. 2). The core equipment is a ceiling-mounted, movable 3.0 T magnet with a 70 cm working aperture. Prior to the iMRI device being moved into the operating room, the diagnostic room was closed for 30 minutes of laminar flow for sterilization. All of the instruments and tables, except for the MRI-safe subcutaneous needle electrodes (Supplementary Fig. 3A), were withdrawn by the nurses beyond the 5Gauss boundary line. To facilitate airway management and to make the awake craniotomy much less challenging in the movable MRI environment, we introduced the Mayo Clinic’s minimal draping25 technique to the IMRIS system (Supplementary Fig. 3B–D). After the MRI safe checklist was completed, the iMRI device was introduced into the operating room, and an interdissection MRI was obtained (Supplementary Fig. 3E and F). The sequences of structural MRI were the same as the preoperative ones. Once scanning was finished, the drape and the adhesive clear plastic drape were re-
moved, and the wound was redraped. If a residual tumor that was amenable to further resection was identified (Fig. 2), the navigational images were updated for further resection guidance. Because MRI-safe subcutaneous needle electrodes were used, brain mapping could be conducted again. Intraoperative scanning was repeated until the iMRI confirmed that maximal tumor resection had been achieved. 2.6. Postoperative language evaluation The same neuropsychologist performed postoperative language assessments before patient discharge (within 7–10 days) and at the 1-month and 3-month follow-up to demonstrate whether surgery resulted in improvement (the increase of AQ > 10 points), deterioration (the decrease of AQ > 10 points), or no change in language functions (increase or decrease 6 10 points) compared with the preoperative baseline evaluation. Patients who had postoperative impairment that had not recovered at 3 months were tested again at 6 months. 2.7. Control of volumetric resection and statistical analysis For contrast-enhancing gliomas, 3D MPRAGE images with gadolinium were used for the volume determination before and after resection. FLAIR imaging was used for nonenhancing gliomas. A postoperative MRI was obtained within 72 hours of the craniotomy procedure in those patients who had further resections but no iMRI confirmation. If there was no further resection, the final iMRI scan
Fig. 2. A 30-year-old man with a left frontal insular astrocytoma (World Health Organization Grade II). Axial fluid-attenuated inversion recovery (FLAIR) images showing tumor volumes. The preoperative (Pre-op) image shows the initial tumor volume to be 87.9 cm3. Intraoperative (Intra-op) image detected a residual tumor with a volume of 21.9 cm3, and the postoperative (Post-op) image obtained after additional resection confirmed a small remnant (6.0 cm3) at the posterior resection border. At the 1-month follow up, the remnant was stable. However, at the 3-month follow up, the residual tumor had disappeared after radiotherapy, and the patient exhibited no language deficits.
J. Lu et al. / Journal of Clinical Neuroscience 20 (2013) 1280–1287
1283
Fig. 3. A 40-year-old man with a left parietal glioblastoma multiforme (World Health Organization Grade IV). (A–D) Transaxial T1-weighted contrast-enhanced images. (A) Preoperative image showing the initial tumor (34.9 cm3). (B) Intraoperative image showing no residual tumor (final tumor volume: 0 cm3). The 1-month (C) and 3-month (D) follow-up images showing no tumor progression. No transient or permanent language deficits occurred. Photographs showing the intraoperative view (E) before and (F) after the resection performed awake under local anesthesia. (G) Schematic drawing showing the relationships between stimulated positive sites and the tumor. The eloquent cortical sites are marked by numbers or letters as follows: L: language cortex inducing speech arrest during counting, F: primary motor cortex of the face, H: primary motor cortex of the hand. (This figure is available in colour at www.sciencedirect.com.)
was considered the postoperative MRI (Fig. 3). In this cohort, contrast enhancement was shown in all high-grade gliomas (HGG) and was not shown in any low-grade gliomas (LGG). All pre- and postoperative tumoral segmentations were manually performed using the OsiriX (Pixmeo, Bernex, Switzerland) software tool by a neurosurgeon and were verified by an additional neurosurgeon.30 The EOR was calculated as follows: (preoperative tumor volume - postoperative tumor volume)/preoperative tumor volume.5 The patients were divided into two groups: patients without further tumor resection after iMRI (group 1) and patients who underwent the removal of additional tumor tissue after iMRI (group 2). For each group, we subdivided the members into the LGG (World Health Organization [WHO] Grade II) and HGG (WHO Grades III and IV) subgroups.31 In patients who underwent further tumor tissue removal (group 2), the final EOR was compared with the EOR obtained from the first intraoperative scan by the Wilcoxon signed-rank test, and a significance level of 0.05 was assumed (p < 0.05). Statistical analyses were performed using STATA software (version 10, StataCorp LP, College Station, Texas, USA). In this study, gross total resection (GTR) was defined as 100% resection of the tumor volume. 3. Results From March 2011 to November 2011, 39 patients were assessed for eligibility. Of these, two patients did not meet the inclusion criteria. Six patients were excluded because of intracranial hypertension (n = 5) and severe language deficits (n = 1). One patient refused to participate in the study. In total, 30 consecutive patients (21 men and nine women), with a median age of 45.5 years (range, 19–67 years), were prospectively evaluated (Table 1). The most common histopathologic diagnosis was diffuse astrocytoma (n = 13, 43.3%), followed by glioblastoma multiforme (n = 10, 33.3%), oligodendroglioma (n = 4, 13.3%), oligoastrocytoma (n = 2,
6.7%) and anaplastic oligodendroglioma (n = 1, 3.3%). Newly diagnosed gliomas were treated in 25 procedures, and recurrent gliomas were treated in five. The majority of the tumors were located in the left frontal lobe (n = 14, 46.7%), followed by the insular lobe (n = 8, 26.7%), the parietal lobe (n = 4, 13.3%) and the temporal lobe (n = 4, 13.3%). 3.1. Extent of resection In this study, all of the surgical procedures were performed by consultant neurosurgeons (J. Wu and Y. Mao). The first-look iMRI confirmed GTR in 11 (36.7%) patients. Of the 19 patients in which complete resection was not achieved as determined by the firstlook iMRI, further resection was not attempted in eight patients because of tumor involvement in the eloquent brain areas. Overall, GTR was achieved in 18 (60.0%) of the 30 patients. Thus, the use of iMRI increased the GTR rate from 36.7% to 60.0%, which represents a 23.3% increase. When the EOR was stratified according to tumor grade, GTR was achieved in 47.4% of the LGG and 81.8% of the HGG. Tables 2 and 3 list the volumetric assessment in the two groups, patients without further resection (group 1) and patients with further resection (group 2). For the 12 patients with LGG in group 1 (Table 2), the median EOR was 96.3% (range, 73.9–100%), and GTR was accomplished in five patients. Six of seven patients with HGG achieved GTR. Among the 19 patients in group 1, the median pre- and postoperative tumor volumes were 62.3 cm3 (range, 8.4–216.7 cm3) and 0 cm3 (range, 0–56.5 cm3), respectively. The median EOR for all 19 patients was 100% (range, 73.9–100%). The gliomas in group 2 (Table 3) included seven LGG and four HGG that underwent further resection. Because iMRI was used, GTR was accomplished in seven of 11 patients. Remnants were present in four patients (three LGG and one HGG) after additional tumor resection as a result of the invasion of critical structures. In
1284
J. Lu et al. / Journal of Clinical Neuroscience 20 (2013) 1280–1287 Table 1 Demographics of patient undergoing tumor resection with awake language maping and iMRI Variable
Value
Age (years) Median Range
45.5 19–67
Male sex: No. (%)
21 (70.0)
Signs and symptoms at presentation: No. (%) Seizure Headache Language deficit Paresthesia
16 (53.3) 9 (30.0) 7 (23.3) 1 (3.3)
WHO tumor grade and histologic type: No. (%) II, diffuse astrocytoma II, oligodendroglioma II, oligoastrocytoma III, anaplastic oligodendroglioma IV, glioblastoma multiforme
13 (43.3) 4 (13.3) 2 (6.7) 1 (3.3) 10 (33.3)
Tumor location: No. (%) Frontal Parietal Temporal Insular
14 (46.7) 4 (13.3) 4 (13.3) 8 (26.7)
Tumor volume (cm3) Median Range
60.0 8.4–216.7
Gross total resection according to WHO tumor grade31 (in the first iMRI): No./total No. (%) All grades I or II III or IV
11/30 (36.7) 5/19 (26.3) 6/11 (54.5)
Gross total resection according to WHO tumor grade: No./total No. (%) All grades I or II III or IV
18/30 (60.0) 9/19 (47.4) 9/11 (81.8)
New or increased postoperative language deficit: No./total No. (%) 1 week 1 month 3 months 6 months
12 (40.0) 5 (16.7) 2 (6.7) 1 (3.3)
iMRI = intraoperative MRI, WHO = World Health Organization.
Table 2 Summary of volumetric assessment in 19 patients without further tumor resection after iMRI Tumor grade
No. of patients
Median initial tumor volume cm3 (range)
Median residue volume on iMRI cm3 (range)
Median EOR (range)
No. of patients with GTR
LGG HGG Total
12 7 19
68.0 (9.1–216.7) 57.0 (8.4–81.9) 62.3 (8.4–216.7)
2.7 (0–56.5) 0 (0–3.7) 0 (0–56.5)
96.3% (73.9–100%) 100% (95.5–100%) 100% (73.9–100%)
5 6 11
EOR = extent of resection, GTR = gross total resection, HGG = high-grade glioma, iMRI = intraoperative MRI, LGG = low-grade glioma.
the LGG subgroup, the final EOR were significantly higher compared with those obtained from the first iMRI (p = 0.02). There was no significant difference between the EOR determined from the first iMRI and the final EOR (p = 0.07) in the HGG subgroup, although the median EOR increased from 95.8% (range, 82.8– 96.3%) to 100% (range, 94.2–100%). Among the 11 patients, the median EOR increased from 92.5% (range, 75.1–97.0%) to 100% (range, 92.6–100%), which was statistically significant (p < 0.01).
follow up, five (16.7%) patients had not returned to their baseline function. By 3 months, only two (6.7%) of the 30 patients had new language deficits (anomia). At the 6-month follow up, only one (3.3%) had permanent language deficits (Table 4).
3.2. Functional outcome
Although four patients presented with focal seizures (myoclonic twitches) during mapping, cold saline was able to stop the attacks. There were no generalized seizures during any of the procedures. There were no complications related to the iMRI or any adverse events caused by the high-field MRI. All of the patients tolerated the processes well and without complaints of intraoperative moderate or severe pain.32
During the preoperative language assessment, seven patients had language deficits that were not significant enough to prevent them from performing intraoperative tasks. One week after surgery, 12 out of 30 (40.0%) patients (Table 4) demonstrated deterioration (n = 4) or new language deficits (n = 8). At the 1-month
3.3. Complications
1285
J. Lu et al. / Journal of Clinical Neuroscience 20 (2013) 1280–1287 Table 3 Summary of volumetric assessment in 11 patients who underwent removal of additional tumor tissue after intraoperative MRI Tumor grade
No. of patients
Median initial tumor volume cm3 (range)
Median residue volume in the first iMRI cm3 (range)
Median EOR in the first iMRI (range)
Median final tumor volume cm3 (range)
Median final EOR (range)
pvalue
No. of patients with GTR (%)
LGG HGG Total
7 4 11
57.7 (35.8–103.9) 59.8 (29.1–80.9) 57.7 (29.1–103.9)
5.7 (2.1–21.9) 3.0 (2.2–5.0) 3.0 (1.4–21.9)
89.0% (75.1–97.0%) 95.8% (82.8–96.3%) 92.5% (75.1–97.0%)
0 (0–7.7) 0 (0–1.7) 0 (0–7.7)
100% (92.6–100%) 100% (94.2–100%) 100% (92.6–100%)
0.02 0.07 <0.01
4 3 7
EOR = extent of resection, GTR = gross total resection, HGG = high-grade glioma, iMRI = intraoperative MRI, LGG = low-grade glioma.
Table 4 Clinical characteristics and results of 12 patients who suffered from transient deficits following resection for gliomas in language areas Patient No.
Age (years), Sex
Location
6 8 9 10 12 13 14 15 27 28 29 30
66, 49, 41, 47, 40, 48, 27, 63, 65, 58, 46, 36,
T P I P I I I I Fr I T I
M M M M F F M M M M M M
Pathology (WHO grade)31
Initial tumor volume (cm3)
EOR on the first iMRI (%)
Final EOR (%)
Pre-op AQ
Post-op AQ 1 week
1 month
3 months
GBM (IV) GBM (IV) Astrocytoma (II) Anaplastic Oligodendroglioma (III) Astrocytoma (II) Astrocytoma (II) Astrocytoma (II) Astrocytoma (II) GBM(IV) Oligodendroglioma (II) Oligoastrocytoma(II) Astrocytoma (II)
80.9 29.1 98.6 64.7 103.9 48.7 123.9 73.7 50.9 62.3 47.9 96.4
96.3 82.8 89.0 100.0 80.1 88.3 78.3 87.9 95.7 96.5 97.1 93.9
100.0 94.2 98.3 100.0 92.6 100.0 78.3 87.9 100.0 96.5 100.0 93.9
83.2 78.8 96.6 73.9 97.4 97.5 99.4 98.4 88.4 99.0 95.8 98.5
50.9 46.6 46.3 50.6 27.4 35.1 33.9 49.5 66.1 33.0 73.3 77.5
76.2 73.1 93.0 72.0 73.3 90.2 97.6 69.0 76.7 79.6 98.8 84.7
82.3 81.0 91.5 82.5 84.8 88.6 96.6 66.6 93.5 96.6 97.0 93.8
6 months
87.8
54.5
AQ = aphasia quotient, EOR = extent of resection, F = female, Fr = frontal, GBM = glioblastoma multiforme, I = insular, iMRI = intraoperative magnetic resonance imaging, M = male, P = parietal, Post-op = postoperative, Pre-op = preoperative, T = temporal, WHO = World Health Organization.
Table 5 Comparison with previous studies using both high-field intraoperative MRI guidance and awake craniotomy Authors, year of publication
No. of cases
Strength of iMRI
iMRI concepts
iMRI draping protocols
Airway management during iMRI scan
iMRI-safe subcutaneous needle electrodes
Volumetric analysis
Nabavi et al., 200924 Goebel et al., 201023 Weingarten et al., 200926 Parny et al., 201025 Leuthardt et al., 201127 Lu et al. (present study), 2013
38
1.5 T
Fixed magnet
Uncovering the patient’s face (no details)
ND
ND
No
10
1.5 T
ND
No
1.5 T
Trimming drapes hanging below the operating table (no details) Minimal draping
ND
1
No LMA
ND
No
12
1.5 T
ND
No
3.0 T
Standard draping protocol (the head and upper body are wrapped in sterile drapes) Minimal draping
LMA
30
Fixed magnet Fixed magnet Movable magnet Movable magnet
No LMA
Yes
Yes
25
iMRI = intraoperative magnetic resonance imaging, LMA = laryngeal mask airway, ND = no data.
4. Discussion Despite a lack of class I evidence, an increasing number of studies suggest that more extensive surgical resection is associated with longer life expectancy for both LGG and HGG.7,33 However, extensive resection carries the risk of neurologic morbidity, which affects the quality of life and subsequent overall survival.11 In addition, quality of life is currently deemed to be the highest priority in brain tumor management. Thus, the precise localization and preservation of eloquent areas during surgery is vital, particularly the conservation of language functionality, which is one of the most fundamental skills required for social interaction. 4.1. Localization of language areas Several methods have been applied to the localization of language areas. Language functional magnetic resonance imaging
(fMRI) and brain mapping using DCS are the most common techniques to localize language areas. Although fMRI is safe, noninvasive and repeatable, its reliability for localization remains uncertain.34 Previous studies have shown great variability in the sensitivity (59–100%) and specificity (0–97%) of fMRI compared with DCS34–37 and have confirmed via different language tasks that the distribution of language activation is widespread. As the gold standard, intraoperative language mapping by DCS has been shown to be a good predictor of functional recovery in Indo-European languages such as English and French.14,19–21 Nonetheless, Chinese differs notably from alphabetic languages. Whereas alphabetic words have a linear structure made up of letters, Chinese words have a square, nonlinear configuration that uses pictographic characters as basic units.38 Tan et al.39,40 reported that the process of Chinese involves more brain areas than that of English, so they suspected that injury to or surgery of the brain in Chinese-speakers may result in a higher rate of language deficits. In
1286
J. Lu et al. / Journal of Clinical Neuroscience 20 (2013) 1280–1287
this present study, we assessed the functional outcome of glioma surgery on language areas in Chinese speakers using this gold standard. Unfortunately, we had a higher incidence of transient postoperative language deficits (40.0% at 1 week) compared with two of the largest studies describing awake language mapping in English speakers (Sanai et al., 22.4%19 and Kim et al., 36%20). However, the long-term functional outcomes were similar to the above studies (16.7% at 1 month and 3.3% at 6 months for this study). We suspect that this higher incidence in transient deficits can be attributed to differences between the Chinese and English languages. Of course, we cannot draw a firm conclusion because of the small number of patients enrolled. With regard to EOR, in this study, GTR was achieved in 60% of the patients, which did not demonstrate an obvious advantage for iMRI. If P95%20 was used to calculate the GTR in all cases, 73.3% of the patients achieved complete resection, which is higher than in the two previous studies.19,20 If the assumed 90% EOR threshold for LGG were applied,5 the surgical survival advantage improves from 12 (63.2%) to 16 (84.2%) patients in 19 LGG after iMRI. We believe that this difference can be attributed to the contribution of iMRI. 4.2. Contribution of iMRI and volumetric analysis Over the past decade, a growing body of literature has focused on the added value of iMRI. Approximately 12 non-randomized cohort studies have demonstrated that iMRI can increase the EOR for glioblastoma multiforme, which has also been well reviewed by Kubben et al.4 The evidence-based grade for recommendation of the clinical efficiency of iMRI currently stands at level B.4,6 There were no randomized, controlled trials that provided evidence for the implementation of iMRI in glioma surgery until 2011, when Senft et al.41 conducted the first randomized trial investigating the application of iMRI in patients with contrast-enhancing gliomas. They confirmed the beneficial effects of 0.15 T iMRI for GTR compared with conventional microsurgery. The rate of GTR by volumetric analysis in the iMRI group was obviously higher than that in the control group (i.e., 96% versus 68%, p = 0.02). In addition, several non-randomized cohort studies have also investigated the contribution of iMRI to EOR using a volumetric assessment.42–44 Kuhnt et al.44 reported what was previously the largest dataset, which demonstrated that the percentage of complete resections increased from 31.7% to 38.6%. Our data show that the percentage of GTR increased from 36.7 to 60.0%. The final GTR in this study was higher, which was perhaps attributable to the higher number of HGG, which had a higher rate of GTR (81.8%). Furthermore, the majority of the published reports, including volumetric assessments, have also reported rates of continued surgery, ranging roughly from 26% to 53%. Our rate of 36.7% falls within this range. Interestingly, it seems that a higher rate of continued surgery is associated with a higher number of patients achieving GTR. The aforementioned studies included patients with gliomas involving non-eloquent areas, therefore the functional outcomes of our study are not directly comparable.
combination were confirmed by Weingarten et al.26 and minimal draping was reported in the Mayo Clinic’s case report (n = 1).25 There were several features in common in the above-mentioned studies. The initial experiences were all conducted in a 1.5 T MRI environment, and all physicians preferred transporting the patients over moving the MR scanner itself. However, in a recent case study, Leuthardt et al.27 described the integration of awake craniotomy in a movable high-field (1.5 T) MRI environment. The differences between previous studies and the present study are shown in Table 5. The highlighted point is the airway management and the resulting pattern of draping. As Leuthardt et al.27 described, airway management is critical to this combination. The awake anesthesia approach that they adopted was similar to the ‘‘asleepawake-asleep’’ technique,45 which places a laryngeal mask over the airway during scanning. Although this technique can provide a secure airway and hyperventilate the patient during the operation, it involves elaborate intraoperative airway maneuvers and requires a particular head positioning. Furthermore, topical anesthesia during re-intubation of the upper airway may result in the loss of airway reflexes. The MAC46 technique that we adopted during the procedure was comparatively easy to perform and allowed the patient to be awakened at any time. This protocol provided adequate sedation and analgesia without a laryngeal mask airway. Nevertheless, adopting MAC in high-field iMRI units is problematic for standard draping protocols because of airway protection.25,27 The minimal draping technique25 effectively solved these issues and was confirmed as safe, sterile and simple. Hence, we introduced this technique into the movable 3.0 T iMRI environment and validated that it could be safely and effectively applied. 4.4. Study limitations There were some limitations to the current study. First, the small number of patients limited our ability to draw any conclusion on whether the combined procedure affected the eventual outcomes. However, our data could contribute to a growing worldwide database, and larger datasets are being collected. Furthermore, we did not perform language subcortical stimulation in our series because it is time consuming. Neuronavigation of the arcuate fasciculus was applied in the requisite cases. Finally, this study reported a case series, and we did not include comparisons with a control group. To unequivocally demonstrate the value of iMRI, a prospective, randomized, controlled clinical trial is underway at our center.
5. Conclusion In this study, we present our clinical experience with 30 patients who underwent glioma resection aided by the combination of awake language mapping and iMRI guidance. It demonstrates the potential utility of combining awake craniotomy with iMRI. It is safe and reliable to perform awake surgery in a movable iMRI.
4.3. Comparison with previous studies combining iMRI guidance and awake craniotomy
Acknowledgments
Although these two techniques have been widely used, only four neurosurgical units have reported their preliminary clinical experiences with the combined use of awake craniotomy and brain mapping in high-field iMRI surgical suites (Table 5). The first integration of iMRI and awake craniotomy was undertaken by Nabavi et al.24 in 2009. Thereafter, this group also prospectively evaluated 25 patients for depression and anxiety and confirmed that iMRI appears to have no additional significant impact on subjective patient perceptions.23 At the same time, the safety and reliability of this
The authors would like to thank the nurses (Yanwei Gong, Chunmei Li, Ye Wang) and iMRI technician (Zhong Yang) of our team for their perfect cooperation during the complex procedures; Jianbing Shi for technical support of neuronavigation; Geng Xu for intraoperative neurophysiologic support; Qihao Guo and Yan Zhou for the language assessment; Yanyan Song for the statistical analysis and Ian F. Parney for the advice on minimal draping. This work was supported by the Ministry of Health of China (2010–2012), the National Natural Science Foundation of China (Project No.
J. Lu et al. / Journal of Clinical Neuroscience 20 (2013) 1280–1287
81071117) and (XBR2011022).
the
Shanghai
Municipal
Health
Bureau
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jocn.2012.10.042. References 1. Wen PY, Kesari S. Malignant gliomas in adults. N Engl J Med 2008;359:492–507. 2. Claus EB, Horlacher A, Hsu L, et al. Survival rates in patients with low-grade glioma after intraoperative magnetic resonance image guidance. Cancer 2005;103:1227–33. 3. Claus EB, Black PM. Survival rates and patterns of care for patients diagnosed with supratentorial low-grade gliomas: data from the SEER program, 1973– 2001. Cancer 2006;106:1358–63. 4. Kubben PL, ter Meulen KJ, Schijns OE, et al. Intraoperative MRI-guided resection of glioblastoma multiforme: a systematic review. Lancet Oncol 2011;12:1062–70. 5. Smith JS, Chang EF, Lamborn KR, et al. Role of extent of resection in the longterm outcome of low-grade hemispheric gliomas. J Clin Oncol 2008;26:1338–45. 6. Kuhnt D, Becker A, Ganslandt O, et al. Correlation of the extent of tumor volume resection and patient survival in surgery of glioblastoma multiforme with highfield intraoperative MRI guidance. Neuro Oncol 2011;13:1339–48. 7. Sanai N, Berger MS. Glioma extent of resection and its impact on patient outcome. Neurosurgery 2008;62:753–64. 8. Diez Valle R, Tejada Solis S, Idoate Gastearena MA, et al. Surgery guided by 5aminolevulinic fluorescence in glioblastoma volumetric analysis of extent of resection in single-center experience. J Neurooncol 2011;102:105–13. 9. Lacroix M, Abi-Said D, Fourney DR, et al. A multivariate analysis of 416 patients with glioblastoma multiforme: prognosis, extent of resection, and survival. J Neurosurg 2001;95:190–8. 10. Vecht CJ, Avezaat CJ, van Putten WL, et al. The influence of the extent of surgery on the neurological function and survival in malignant glioma. A retrospective analysis in 243 patients. J Neurol Neurosurg Psychiatry 1990;53:466–71. 11. McGirt MJ, Mukherjee D, Chaichana KL, et al. Association of surgically acquired motor and language deficits on overall survival after resection of glioblastoma multiforme. Neurosurgery 2009;65:463–9 discussion 469–470. 12. Lang MJ, Kelly JJ, Sutherland GR. A moveable 3-Tesla intraoperative magnetic resonance imaging system. Neurosurgery 2011;68(1 Suppl Operative):168–79. 13. Nimsky C, Ganslandt O, Von Keller B, et al. Intraoperative high-field-strength MR imaging: implementation and experience in 200 patients. Radiology 2004;233:67–78. 14. Ojemann G, Ojemann J, Lettich E, et al. Cortical language localization in left, dominant hemisphere. An electrical stimulation mapping investigation in 117 patients. J Neurosurg 1989;71:316–26. 15. Duffau H, Capelle L, Sichez N, et al. Intraoperative mapping of the subcortical language pathways using direct stimulations. An anatomo-functional study. Brain 2002;125:199–214. 16. Zhao Y, Chen X, Wang F, et al. Integration of diffusion tensor-based arcuate fasciculus fibre navigation and intraoperative MRI into glioma surgery. J Clin Neurosci 2012;19:255–61. 17. Sun GC, Chen XL, Zhao Y, et al. Intraoperative MRI with integrated functional neuronavigation-guided resection of supratentorial cavernous malformations in eloquent brain areas. J Clin Neurosci 2011;18:1350–4. 18. Nimsky C, Ganslandt O, Hastreiter P, et al. Intraoperative compensation for brain shift. Surg Neurol 2001;56(6):357–64 discussion 364–355. 19. Sanai N, Mirzadeh Z, Berger MS. Functional outcome after language mapping for glioma resection. N Engl J Med 2008;358:18–27. 20. Kim SS, McCutcheon IE, Suki D, et al. Awake craniotomy for brain tumors near eloquent cortex: correlation of intraoperative cortical mapping with neurological outcomes in 309 consecutive patients. Neurosurgery 2009;64:836–45 discussion 345–836. 21. Duffau H, Capelle L, Sichez J, et al. Intra-operative direct electrical stimulations of the central nervous system: the Salpetriere experience with 60 patients. Acta Neurochir (Wien) 1999;141:1157–67.
1287
22. Lu JF, Zhang J, Wu JS, et al. Awake craniotomy and intraoperative language cortical mapping for eloquent cerebral glioma resection: preliminary clinical practice in 3.0 T intraoperative magnetic resonance imaging integrated surgical suite. Zhonghua Wai Ke Za Zhi 2011;49:693–8. 23. Goebel S, Nabavi A, Schubert S, et al. Patient perception of combined awake brain tumor surgery and intraoperative 1.5-T magnetic resonance imaging: the Kiel experience. Neurosurgery 2010;67:594–600 discussion 600. 24. Nabavi A, Goebel S, Doerner L, et al. Awake craniotomy and intraoperative magnetic resonance imaging: patient selection, preparation, and technique. Top Magn Reson Imaging 2009;19:191–6. 25. Parney IF, Goerss SJ, McGee K, et al. Awake craniotomy, electrophysiologic mapping, and tumor resection with high-field intraoperative MRI. World Neurosurg 2010;73:547–51. 26. Weingarten DM, Asthagiri AR, Butman JA, et al. Cortical mapping and frameless stereotactic navigation in the high-field intraoperative magnetic resonance imaging suite technical note. J Neurosurg 2009;111:1185–90. 27. Leuthardt EC, Lim CC, Shah MN, et al. Use of movable high-field-strength intraoperative magnetic resonance imaging with awake craniotomies for resection of gliomas: preliminary experience. Neurosurgery. 2011;69:194–205 discussion 205–196. 28. Gao SR, Chu YF, Shi SQ, et al. A standardization research of the aphasia battery of Chinese. Chinese Mental Health Journal 1992;6:125–8. 29. Kertesz A. The Western Aphasia Battery test manual: New York: Grune & Stratton; 1982. 30. Rosset A, Spadola L, Ratib O. OsiriX: an open-source software for navigating in multidimensional DICOM images. J Digit Imaging 2004;17:205–16. 31. Louis DN, Ohgaki H, Wiestler OD, et al. The 2007 WHO classification of tumours of the central nervous system. Acta Neuropathol 2007;114:97–109. 32. Hartrick CT, Kovan JP, Shapiro S. The numeric rating scale for clinical pain measurement: a ratio measure? Pain Pract 2003;3:310–6. 33. Sanai N, Berger MS. Operative techniques for gliomas and the value of extent of resection. Neurotherapeutics 2009;6:478–86. 34. Giussani C, Roux FE, Ojemann J, et al. Is preoperative functional magnetic resonance imaging reliable for language areas mapping in brain tumor surgery? Review of language functional magnetic resonance imaging and direct cortical stimulation correlation studies. Neurosurgery 2010;66:113–20. 35. Bizzi A, Blasi V, Falini A, et al. Presurgical functional MR imaging of language and motor functions: validation with intraoperative electrocortical mapping. Radiology 2008;248:579–89. 36. Rutten GJ, Ramsey NF, van Rijen PC, et al. Development of a functional magnetic resonance imaging protocol for intraoperative localization of critical temporoparietal language areas. Ann Neurol 2002;51:350–60. 37. Spena G, Nava A, Cassini F, et al. Preoperative and intraoperative brain mapping for the resection of eloquent-area tumors. A prospective analysis of methodology, correlation, and usefulness based on clinical outcomes. Acta Neurochir 2010;152:1835–46. 38. Siok WT, Perfetti CA, Jin Z, et al. Biological abnormality of impaired reading is constrained by culture. Nature 2004;431:71–6. 39. Tan LH, Laird AR, Li K, et al. Neuroanatomical correlates of phonological processing of Chinese characters and alphabetic words: a meta-analysis. Hum Brain Mapp 2005;25:83–91. 40. Tan LH, Spinks JA, Gao JH, et al. Brain activation in the processing of Chinese characters and words: a functional MRI study. Hum Brain Mapp 2000;10:16–27. 41. Senft C, Bink A, Franz K, et al. Intraoperative MRI guidance and extent of resection in glioma surgery: a randomised, controlled trial. Lancet Oncol 2011;12:997–1003. 42. Hatiboglu MA, Weinberg JS, Suki D, et al. Impact of intraoperative high-field magnetic resonance imaging guidance on glioma surgery: a prospective volumetric analysis. Neurosurgery. 2009;64:1073–81 discussion 1081. 43. Schneider JP, Trantakis C, Rubach M, et al. Intraoperative MRI to guide the resection of primary supratentorial glioblastoma multiforme–a quantitative radiological analysis. Neuroradiology 2005;47:489–500. 44. Kuhnt D, Ganslandt O, Schlaffer SM, et al. Quantification of glioma removal by intraoperative high-field magnetic resonance imaging – an update. Neurosurgery 2011;69:852–62 discussion 862–863. 45. Huncke K, Van de Wiele B, Fried I, et al. The asleep-awake-asleep anesthetic technique for intraoperative language mapping. Neurosurgery. 1998;42:1312–6 discussion 1316–1317. 46. Berkenstadt H, Perel A, Hadani M, et al. Monitored anesthesia care using remifentanil and propofol for awake craniotomy. J Neurosurg Anesthesiol 2001;13:246–9.