Abstract BACKGROUND Minimally invasive techniques are increasingly being used to access intra-axial brain lesions. OBJECTIVE To describe a method of resecting frontal gliomas through a keyhole craniotomy and share the results with these techniques. METHODS We performed a retrospective review of data obtained on all patients undergoing resection of frontal gliomas by the senior author between 2012 and 2015. We describe our technique for resecting dominant and nondominant gliomas utilizing both awake and asleep keyhole craniotomy techniques. RESULTS After excluding 1 patient who received a biopsy only, 48 patients were included in the study. Twenty-nine patients (60%) had not received prior surgery. Twenty-six patients (54%) were diagnosed with WHO grade II/III tumors, and 22 patients (46%) were diagnosed with glioblastoma. Twenty-five cases (52%) were performed awake. At least 90% of the tumor was resected in 35 cases (73%). Three of 43 patients with clinical follow-up experienced permanent deficits. CONCLUSION We provide our experience in using keyhole craniotomies for resecting frontal gliomas. Our data demonstrate the feasibility of using minimally invasive techniques to safely and aggressively treat these tumors. Keyhole, Glioma, Craniotomy, Frontal lobe, Resection, Minimally invasive ABBREVIATIONS ABBREVIATIONS DTI diffusion tensor imaging EOR extent of resection IFOF inferior fronto-occipital fasciculus ioMRI intraoperative magnetic resonance imaging LOS length of hospital stay MRI magnetic resonance imaging POD postoperative day SLF superior longitudinal fasciculus Keyhole craniotomies have been employed for treating various intra-axial lesions throughout the brain. However, surgeons often feel that limited-size approaches carry higher risk resulting from limited visualization of critical structures and restricted ability to control intraoperative hemorrhage.1 Despite this, many advances have been made. Suitable access to relevant structures is critical to safe and effective neurosurgery. However, what constitutes adequate visualization remains ill-defined. We would argue that the ideal approach is the least destructive method that can safely and effectively meet the demands of the case being performed. Our goal in glioma surgery is to achieve supramaximal resection as initially described by Duffau.2,3 We demonstrate that this approach can be used for a complete lobectomy; however, it can also be tailored to a more limited resection. Our technique is driven by the premise that most of the risk in a keyhole frontal craniotomy for tumor resection and frontal lobectomy occurs with the posterior disconnection between tumor and eloquence. In the present study, we provide our experience with frontal keyhole craniotomies involving dominant and nondominant low- and high-grade gliomas. With a standardized plan, we achieve resection of tumors in a variety of frontal locations. This feasibility study provides data on the technical aspects of frontal keyhole craniotomies and details outcomes of patients who have received an operation with this minimally invasive method. METHODS Patient Selection We performed a retrospective review of data on all patients undergoing frontal keyhole craniotomy performed by the senior author during a 3-yr period from 2012 to 2015 at our institution. We included all patients with a histopathologic diagnosis of WHO grade 2 or 3 oligodendroglioma/astrocytoma, or glioblastoma who underwent surgical resection. Biopsy-only patients were not included. Clinical records, hospital charts, and imaging studies were reviewed through the last available follow-up. Patients who were not seen at least 3 mo postoperatively were noted as lost to follow-up. Medical history, operative notes, and hospital course were also reviewed. Length of hospital stay (LOS) was noted beginning with the day of the operation and ending the day of discharge. This review was performed with approval of our institutional review board. Preoperative Assessment Magnetic resonance imaging (MRI) with and without gadolinium contrast was performed preoperatively in all patients. Using diffusion tensor imaging (DTI) tractography, the corticospinal tract, superior longitudinal fasciculus (SLF), arcuate fasciculus, inferior fronto-occipital fasciculus (IFOF), and optic pathways were included into image guidance for operative planning and resection, as described elsewhere.4 Patients underwent preoperative clinical evaluation by physical and speech therapists, including spatiotemporal, attention, baseline speech, and motor testing. Awake surgery was performed in cases where the tumor was in proximity of eloquent structures in the frontal lobe. These included premotor and motor planning areas (ie, supplementary motor cortex SMA), as well as the IFOF and SLF. Awake surgery was employed more frequently in resecting tumors throughout the frontal lobe over time, especially the middle frontal lobe. Patients were consented for surgery after reviewing the risks and alternatives. Awake Mapping Technique We employ awake mapping in any patient where tumor may involve an eloquent part of the brain. Awake keyhole craniotomies utilize negative mapping exclusively, and a negative site is confirmed by at least 3 separate stimulations at a no greater than 6 mA (average is 4 mA). This appears to be a high enough current to confirm truly negative sites; however, low enough to avoid after discharges and seizure activity. If a site at the planned disconnection is confirmed negative, then it is included in the resection. Some patients are unable to undergo awake surgery, including those with severe anxiety arising from the prospect of having awake craniotomy performed, and also those with moderate baseline confusion and inability to complete preoperative task assessment reliably. In these instances, diffusion tractography is utilized to guide the resection without the confirmatory aid of negative mapping. Surgical Technique The posterior disconnection is planned anterior to the premotor/motor areas as well as the SLF. Prior to disconnection, awake cortical, followed by subcortical mapping is carried out; typically involving patients performing contralateral motor tasks as well as picture naming for the cortical mapping portion. As subcortical dissection is carried out, the patient is instructed to perform “double-tasking,” ie, naming and contralateral motor movement, continuously until the tumor-eloquence disconnection is completed. In our experience “double-tasking” increases the sensitivity of identification of eloquent regions, demonstrating subtle deficits earlier than without simultaneous tasking. Typically, total disconnection for this cut is considered when the anterior circular sulcus of the insula is identified. Further risk is incurred in the dominant hemisphere when making the lateral disconnection, which abuts the SLF superficially. Hence, the keyhole craniotomy is planned over the posterior and lateral aspect of the tumor frontal cortex boundary at the planned division site using image guidance (Figures 1A and 1B). When pertinent, these eloquent cortices are then mapped awake and are thus preserved during the tumor resection in the same fashion that the posterior cut is completed. The lateral cut is considered complete when the ipsilateral IFOF is identified. The deep cut, which is intended to preserve the IFOF, is considered complete once the frontal horn is identified. Thus, this division tracks lateral-to-medial in order to preserve the laterally predominant SLF, with a trajectory ending near the caudate/frontal horn. Given that the bone flap does not completely expose the lateral frontal cortex, some lateral cortical mapping is limited initially. However, as the lateral division is created, subcortical mapping allows “inside-out” resection, preserving the SLF and its cortical input sites. The rationale behind limited lateral cortical exposure is based on the configuration of the SLF, which in most patients is very large. Damage to this tract, specifically on the dominant hemisphere, is avoided. FIGURE 1. View largeDownload slide Craniotomy planning based upon the tumor/brain interface with overlay fiber tractography. Schematic 3-D representation of white matter tracts and the tumor with a T1-weighted MRI for orientation. A, Axial representation. The craniotomy that is planned with image guidance allows for a posterior disconnection from the corticospinal tract and SLF and a lateral disconnection from the SLF. The planned craniotomy is outlined with white dash marks. It is specifically planned to expose the tumor/cortical–subcortical interface. B, Sagittal representation. The craniotomy plan, shown with white dash marks, underscores the lateral extent in relation to the SLF, which is pushed inferiorly. As described, subcortical negative mapping in this context allows identification and thus avoidance of the SLF from the interior of the resection cavity as opposed to reliance on cortical negative mapping. The cortical interface is also negatively mapped on the lateral surface of the exposed craniotomy. Corticospinal tract: blue; SLF: orange; IFOF: pink; optic pathways: green; tumor: red. FIGURE 1. View largeDownload slide Craniotomy planning based upon the tumor/brain interface with overlay fiber tractography. Schematic 3-D representation of white matter tracts and the tumor with a T1-weighted MRI for orientation. A, Axial representation. The craniotomy that is planned with image guidance allows for a posterior disconnection from the corticospinal tract and SLF and a lateral disconnection from the SLF. The planned craniotomy is outlined with white dash marks. It is specifically planned to expose the tumor/cortical–subcortical interface. B, Sagittal representation. The craniotomy plan, shown with white dash marks, underscores the lateral extent in relation to the SLF, which is pushed inferiorly. As described, subcortical negative mapping in this context allows identification and thus avoidance of the SLF from the interior of the resection cavity as opposed to reliance on cortical negative mapping. The cortical interface is also negatively mapped on the lateral surface of the exposed craniotomy. Corticospinal tract: blue; SLF: orange; IFOF: pink; optic pathways: green; tumor: red. We ensure that the medial edge of the craniotomy is roughly 1.5 cm from the lateral edge of the sagittal sinus to facilitate safe reach under the bone flap to address the medial frontal lobe. The anterior limit of the bone flap is near the beginning of the frontal boss to accommodate anterior visualization of the superior frontal lobe. Cortical mapping far anterior (ie, the frontal pole) to the planned posterior division is unnecessary, as the posterior cut disconnects the more anterior region of the cortical and subcortical network; which emphasizes the importance of preoperative planning with respect to the limits of the SLF and SMA. Once the craniotomy is planned, a linear incision of 4 to 5 cm is made in the sagittal plane. The craniotomy is approximately 2.5 cm in diameter, as shown in Figure 2. The vast majority of the resection is done with the microscope to allow for light and visualization through the smaller opening. The posterior and lateral disconnections are made first, separating eloquent white matter tracts from tumor. The lateral cut is wedge-shaped and proceeds downward medial to SLF until reaching IFOF. The dissection continues medially into the frontal horn, facilitating early identification and preservation of the caudate. In awake surgery, these cuts are monitored with subcortical speech and motor mapping. The anterior cerebral artery complex is separated from the tumor using the subpial technique. Near the midline, the ipsilateral cingulate gyrus is preserved if not involved with tumor, aided by early identification of the cingulate sulcus. If tumor extension exists through the corpus callosum, this is followed underneath the contralateral cingulate gyrus. Once completed, we continue the posteromedial cut until it is in front of the caudate head. The orbitofrontal cortex, unless involved with tumor, is typically preserved deep and anterior to the resection. However, if resection is indicated given the extent of tumor involvement, this approach easily allows complete resection of this portion of the frontal lobe. The frontal pole is freed from bridging veins, folded inward, and removed en bloc under the bone flap. FIGURE 2. View largeDownload slide Frontal keyhole approach. A, The patient is positioned supine with planned craniotomy over the posterior-most involved portion of the frontal lobe as observed with image guidance. Note the craniotomy is not extended to the forehead. B, Planned incision measuring 4 to 5 cm in diameter. C, The craniotomy is approximately 2 to 3 cm in diameter. D, “Keyhole” view down the anterior axis of the frontal lobe. FIGURE 2. View largeDownload slide Frontal keyhole approach. A, The patient is positioned supine with planned craniotomy over the posterior-most involved portion of the frontal lobe as observed with image guidance. Note the craniotomy is not extended to the forehead. B, Planned incision measuring 4 to 5 cm in diameter. C, The craniotomy is approximately 2 to 3 cm in diameter. D, “Keyhole” view down the anterior axis of the frontal lobe. Surgical Technique—Repeat Operations In repeat operations for frontal gliomas, the previous incision design is taken into account. We may alter the standard linear incision (ie, transition from a sagittal incision to a coronal based incision) to avoid further disruption of the scalp blood supply. The bone flap design is relatively consistent between a new case and a repeat operation. Again, the preoperative planning phase utilizing fiber tractography is employed to position the craniotomy over the intended surgical trajectory. Repeat craniotomies done on patients who have had a previous surgery at other institutions can be somewhat challenging to plan. Outcome Assessment Postoperatively, patients were evaluated by ancillary support services (physical therapy and speech and language pathology), preoperative task assessment (naming, motor function, etc.) was compared to postoperative task assessment. Patients underwent a full neurological examination by the attending neurosurgeon immediately after surgery and at follow-up in clinic within 3 mo of surgery. Complications were thus noted either at postoperative examination or in clinic. A complication reported following surgery that had resolved by clinic follow-up was recorded as temporary. Surgery duration was recorded from skin opening to skin closure. Tumor volumes were calculated using preoperative postcontrast T1-weighted MRIs by the first author and confirmed by the senior author. For nonenhancing tumors, T2-weighted MRIs were used. We feel that T2-weighted imaging gives a more detailed picture of tumor infiltration when compared to FLAIR sequences. ImageJ (National Institutes of Health) was used for tumor segmentation, as performed by others.5 The tumor was outlined on individual slices by freehand. The volume was calculated as the sum of the individual areas times the slice thickness. Residual tumor was calculated in the same manner, tracing areas of residual contrast enhancement or areas of T2 hyperintensity present in postoperative imaging. Distances were standardized for each imaging study to account for possible variability between baseline image sizes. The extent of resection (EOR) was calculated as follows: (preoperative tumor volume − postoperative tumor volume)/preoperative tumor volume, and recorded within the appropriate percentage range. Statistical Analysis Continuous variables (EOR and time from first case in the series) were compared using linear regression. Probability values greater than .05 are reported as nonsignificant. All data analysis was performed with SPSS (version 22, IBM Inc, Armonk, New York). RESULTS Patient Population A total of 48 patients with WHO grade II to IV gliomas were treated with frontal keyhole craniotomy by the senior author between 2012 and 2015. Characteristics of these patients are given in Table 1. Twenty-two of 48 (46%) patients were treated for glioblastoma and 26 of 48 (54%) were treated for oligodendroglioma or astrocytoma. Twenty-nine patients (60%) had not received a previous operation. Of the 19 patients undergoing a repeat operation, 16 (27%) had received previous operations at other hospitals. A total of 6 patients (12%) underwent more than 1 surgery at our institution for tumor recurrence, including 3 who had also received a surgery at an outside hospital. Three patients underwent repeated operations at our hospital that were completion lobectomies after less-extensive first surgeries. In those cases, the first surgeries were limited because pathology was unknown or because there was residual tumor. Five of 48 (10%) underwent chemotherapy and/or radiation prior to operative resection. Tumor and operative data are given in Table 2. All patients had tumors of the frontal lobe; 29/48 (60%) were left-sided, 18/48 (38%) were right-sided, and 1/48 (2%) was bilateral. The median tumor volume was 20.6 cm3, with the largest tumor resected having a volume of 137.6 cm3. TABLE 1. Patient Characteristics Patients n Median age (range) Total 48 46 (24-76) Women 24 41 (24-68) Men 24 52 (27-76) Pathology n Glioblastoma 22 (46%) Oligodendroglioma 15 (31%) Astrocytoma 11 (23%) Tumor grade n IV 22 (46%) II 18 (37%) III 8 (17%) Surgical history n No prior surgery 29/48 (60%) Outside surgery 16/48 (33%) Reoperation 6/48 (12%) Treatment prior to surgery n None 43 (90%) Radiation 2 (4%) Chemo and radiation 2 (4%) Chemotherapy 1 (2%) Patients n Median age (range) Total 48 46 (24-76) Women 24 41 (24-68) Men 24 52 (27-76) Pathology n Glioblastoma 22 (46%) Oligodendroglioma 15 (31%) Astrocytoma 11 (23%) Tumor grade n IV 22 (46%) II 18 (37%) III 8 (17%) Surgical history n No prior surgery 29/48 (60%) Outside surgery 16/48 (33%) Reoperation 6/48 (12%) Treatment prior to surgery n None 43 (90%) Radiation 2 (4%) Chemo and radiation 2 (4%) Chemotherapy 1 (2%) View Large TABLE 2. Tumor Characteristics and Operative Data Tumor location Left side 29 (60%) Right side 18 (38%) Bilateral 1 (2%) Tumor volume (cm3) Median 20.6 Range 3.5-137.6 Standard dev. 36.5 Mapping Awake surgery Total patients 25 (52%) Asleep surgery Noneloquent area 20 (42%) Could not tolerate awake 3 (6%) Surgery Duration (min) Median awake 213 Median asleep 220 P 0.87 Range 117-568 LOS (days) Mediana 3 Ranges 0-2 19 (40%) 3-6 16 (33%) 7-14 12 (25%) >14 1 (2%) Tumor location Left side 29 (60%) Right side 18 (38%) Bilateral 1 (2%) Tumor volume (cm3) Median 20.6 Range 3.5-137.6 Standard dev. 36.5 Mapping Awake surgery Total patients 25 (52%) Asleep surgery Noneloquent area 20 (42%) Could not tolerate awake 3 (6%) Surgery Duration (min) Median awake 213 Median asleep 220 P 0.87 Range 117-568 LOS (days) Mediana 3 Ranges 0-2 19 (40%) 3-6 16 (33%) 7-14 12 (25%) >14 1 (2%) LOS: Length of hospital stay. aFor all patients, excluding 1 patient who remained in the hospital several weeks for other, unrelated procedures. View Large Operative Results Twenty-five of 48 (52%) patients underwent awake surgery. Twenty patients (42%) were deemed not to require awake surgery based on tumor location, and 3 (6%) were unable to tolerate awake surgery. Median surgery duration was 213 min for patients undergoing awake surgical resection, and 220 min for patients undergoing asleep surgical resection (P = .87). An illustrative case is shown in Figure 3. Median LOS was 3 d. Nineteen of 48 (40%) of patients were discharged on postoperative day (POD) 0 to 2, 16/48 (33%) were discharged on POD 3 to 6, 12/48 (25%) were discharged on POD 7 to 14, and 1/48 (2%) was discharged on POD 14 to 30. FIGURE 3. View largeDownload slide Illustrative case 1. A, Axial T2-weighted MRI demonstrates a hyperintense lesion of the right frontal lobe. B and C, Postoperative T1-weighted MRI with contrast demonstrates resected area in axial B and sagittal C planes. The posterior margin of the craniotomy is observed directly above the posterior margin of the resection in C. FIGURE 3. View largeDownload slide Illustrative case 1. A, Axial T2-weighted MRI demonstrates a hyperintense lesion of the right frontal lobe. B and C, Postoperative T1-weighted MRI with contrast demonstrates resected area in axial B and sagittal C planes. The posterior margin of the craniotomy is observed directly above the posterior margin of the resection in C. EOR ranges are noted in Table 3. Gross total resection was achieved in 24/48 (50%). In 35/48 (73%) patients, 90% to 100% of tumor was resected; in 8/48 (17%) patients, 70% to 89% of tumor was resected; and in 5/48 (10%) patients, 40% to 69% of tumor was resected. Median EOR was 98%. There was no correlation between EOR and experience by the senior author (MES) over time (P = .36) TABLE 3. Outcomes EOR tumora 90%-100% 35 (73%) 70%-89% 8 (17%) 40%-69% 5 (10%) Median 98% GTR 24 (50%) EOR vs surgeon experienceb P = .36 Follow-up Median follow-up (months) 11.5 Range (months) 2-47 Did not follow-up 5 Complicationsc Temporary Permanent Aphasia 3/48 (6%) 2/43 (4%) AMS 2/48 (4%) 0/43 (0%) Paresis 1/48 (2%) 1/43 (2%) Infection 1/48 (2%) – CSF leak 0/48 (0%) 0/43 (0%) Hemorrhage/stroke 0/48 (0%) 0/43 (0%) EOR tumora 90%-100% 35 (73%) 70%-89% 8 (17%) 40%-69% 5 (10%) Median 98% GTR 24 (50%) EOR vs surgeon experienceb P = .36 Follow-up Median follow-up (months) 11.5 Range (months) 2-47 Did not follow-up 5 Complicationsc Temporary Permanent Aphasia 3/48 (6%) 2/43 (4%) AMS 2/48 (4%) 0/43 (0%) Paresis 1/48 (2%) 1/43 (2%) Infection 1/48 (2%) – CSF leak 0/48 (0%) 0/43 (0%) Hemorrhage/stroke 0/48 (0%) 0/43 (0%) AMS: altered mental status; CSF: cerebrospinal fluid; EOR: extent of resection; GTR: gross-total resection. aPostoperative imaging was not available for 1 patient. bDetermined from time since first case included in this series. cSome patients had more than 1 complication. View Large Five patients who missed postoperative clinic appointments and could not be contacted for rescheduling were lost to follow-up (10%). Median follow-up time for all patients was 11.5 mo. Eight patients (17%) experienced postoperative complications. Five patients experienced early aphasia following surgery, which was permanent in 2 cases (7% of left-sided tumors). Postoperative edema and mild cortical damage were thought to be the causes of temporary deficits, whereas damage to white matter structures, such as the IFOF, in the absence of awake mapping, was suspected to be the cause of observed permanent deficits. There were no recorded postoperative hemorrhages or strokes. Other postoperative complications are noted in Table 3. DISCUSSION To date, there has been no large-scale trial comparing keyhole surgery to traditional approaches. However, there is evidence in the form of smaller, focused case series supporting the benefits of keyhole surgery,6-9 particularly involving the supraorbital approach.10 Further, it stands to reason that reduced brain retraction and exploration, improved recovery times, and smaller surgical wounds are in the best interest of the patient.11 The goal of minimally invasive neurosurgery is to achieve maximal efficiency with minimal trauma to the patient. As conceptualized by Perneczky,12 the real benefit of minimally invasive neurosurgery to patients is in limiting trauma that results from exploration of the surgical site and brain retraction. Any keyhole brain surgery is thus more a philosophy than a set of specifically-defined parameters. However, we have outlined the typical incision and bone flap design utilized for this approach at our institution to illustrate the keyhole concept. That is to say, there is no size above which an opening is or is not a keyhole; rather, surgeons make their openings as small as possible to achieve their goal.13 The result is a smaller craniotomy that minimizes brain exposure, likely adding the benefit of early discharge,14 and quicker postoperative recovery. Additionally, faster recovery times can foster earlier initiation of adjuvant therapies such as radiation and chemotherapy. Tumors are essentially resected through keyhole craniotomies by placing the keyhole over the main area of interest, allowing for resection of the tumor along the long axis. In glioma resection, the area of interest is defined as the tumor/eloquent cortex/subcortex boundary. Thus, our craniotomies are planned over the postulated interface of eloquent cortex with tumor. This is accomplished during preoperative planning by utilizing fiber tractography, which in our experience accurately predicts the locations of eloquent regions.5 In this manner, the bone flap does not need to be taken to the full breadth of the resection. Further, we have shown that excellent EOR with low complications can be undertaken utilizing this concept, even near eloquent parenchyma. Types of keyhole craniotomies described include supraorbital, minipterional, occipital transtentorial, suboccipital retrosigmoid, and supracerebellar transtentorial, among others.15-20 They have been employed for epilepsy surgery, aneurysm repair, vascular malformations, and resection of metastatic and primary brain tumors.6,7,21-26 Virtually any intracranial pathology can be approached with the keyhole philosophy. The frontal approach has been a mainstay in our treatment of glioma patients when possible, as surgical resection is the recommended treatment for both low- and high-grade gliomas,27-29 and given the likely benefit of reoperation in select patients.30,31 Frontal gliomas are often treated with awake craniotomies utilizing cortical and subcortical mapping with direct electrical stimulation,32,33 particularly for tumors in eloquent areas.34 Survival has been shown to correlate with EOR in both low- and high-grade glioma,35-37 and consequently DES is used in conjunction with preoperative DTI to safely maximize EOR. Some have also suggested additionally using preoperative transcranial magnetic stimulation, intraoperative magnetic resonance spectroscopy, and intraoperative magnetic resonance imaging (ioMRI) to further ensure maximum safe resection.38-42 Use of any of these modalities is readily accomplished with a keyhole craniotomy, and with adequate anatomic knowledge and careful planning a maximal EOR can also be safely achieved. We preferentially employ awake mapping in any patient where tumor may involve an eloquent part of the brain. Our philosophy evolved with experience, resulting in more cases performed awake. Initially, limited mapping was performed in the frontal lobe, but we adapted our technique over time to include mapping of the middle frontal lobe in light of better understanding of functional networks. In other words, awake surgery was not offered to some patients early in the study who may have been felt to benefit from it later on. Over time, our aim in many cases became to preserve neurological functioning not recorded by standard metrics, focusing on tasks that are not purely speech or motor based. Thus, differences in traditional neurological outcomes presented in this study are likely not all encompassing. Further, we did not initially resect nondominant frontal gliomas using awake surgery as we later did (especially with regard to identifying motor-planning areas, avoiding SMA syndrome, and preserving cingulum anatomy33). It is now standard at our institution to offer awake surgery for nondominant frontal gliomas. In general, we err on the side of caution (viewing awake surgery as the safest with respect to preserving function), electing to do awake cases if there is any likelihood of encountering eloquent tissue intraoperatively. However, in cases where there is little to no possibility of involving eloquent tissue, the operation is performed under general anesthesia to avoid placing the patient under any undue duress. We show the plausibility of frontal keyholes for the resection of gliomas by demonstrating EOR percentages similar to those given in other studies, and emphasize these results can be achieved by careful planning, aided by the use of intraoperative monitoring. We achieved a median EOR of 98%, comparable to 87% to 92% in other studies using mapping and ioMRI.14,23 Tumor size did not preclude the use of this technique, as very large tumors can be fully resected through keyhole craniotomies (tumor size range: 3.5-137.6 cm3). Median tumor volumes recorded in our study are comparable to those noted in previous studies.43,44 The limiting factor in some cases was encountering areas involved in motor, speech, or concentration as identified with awake mapping. In other cases, tumor was residual. An example is given in Figure 4. While experience is likely to increase efficiency in keyhole surgery, a baseline understanding of the unusual angles and operative workflow is paramount. Consequently, statistical analysis demonstrated no significant relationship between EOR (P = .36) and gained operative experience over time. Further insight into complication rates over time would also provide insight into the importance of surgeon experience with these techniques. However, as there were only 3 permanent complications in this series, no conclusions can be drawn about the impact of surgeon experience on complication occurrence. FIGURE 4. View largeDownload slide Illustrative case 2. EOR is limited by one of 2 factors: (1) tumor involvement in eloquent brain tissue; or (2) missing tumor because of limited access/visualization. In our experience, resection is nearly always limited by (1). This imaging illustrates a case of a large glioblastoma of the frontal lobe measuring 5.4 × 8.0 × 6.2 cm, that could not be fully resected due to invasion of eloquent areas. During intraoperative testing, the patient began to have speech difficulty, and resection was halted. Residual tumor resulted from diffuse infiltration of eloquent cortex rather than from limitations of the technique. A-C, FLAIR-weighted axial MRI showing a large hyperintense space-occupying lesion of the left frontal lobe causing compression across the midline. D and E, T1-weighted coronal MRI with contrast from the same patient with rim-enhancement and areas of hypointensity throughout the left frontal lobe. Axial F-H and coronal I, J images from postoperative T1 MRI show frontal lobectomy and resection of tumor. Areas of contrast enhancement suggest residual tumor and are indicated by blue arrows. FIGURE 4. View largeDownload slide Illustrative case 2. EOR is limited by one of 2 factors: (1) tumor involvement in eloquent brain tissue; or (2) missing tumor because of limited access/visualization. In our experience, resection is nearly always limited by (1). This imaging illustrates a case of a large glioblastoma of the frontal lobe measuring 5.4 × 8.0 × 6.2 cm, that could not be fully resected due to invasion of eloquent areas. During intraoperative testing, the patient began to have speech difficulty, and resection was halted. Residual tumor resulted from diffuse infiltration of eloquent cortex rather than from limitations of the technique. A-C, FLAIR-weighted axial MRI showing a large hyperintense space-occupying lesion of the left frontal lobe causing compression across the midline. D and E, T1-weighted coronal MRI with contrast from the same patient with rim-enhancement and areas of hypointensity throughout the left frontal lobe. Axial F-H and coronal I, J images from postoperative T1 MRI show frontal lobectomy and resection of tumor. Areas of contrast enhancement suggest residual tumor and are indicated by blue arrows. With respect to supraresection of glioma tumors, there is some evidence to support an increase in overall survival and progression-free survival.45 Advancements in tumor biology research are likely to continue to shape goals for resection for each individual patient, and current gaps in knowledge about the disease process lend to resection at the discretion of the surgeon. Ultimately, a patient's goals for level of functioning and survival must be carefully considered when planning any operation. Our study additionally demonstrates the safety of keyhole craniotomy in experienced hands. Few patients experienced postoperative neurological deficits, and permanent deficits occurred in only 3/43 (7%) patients. Rates of other complications including infection and cerebrospinal fluid leak were within or below ranges published in earlier studies.34,43 Safety of our technique can be attributed, in part, to separate advances made in the field over the last 20 yr in hemostasis and image guidance. After all, the traditional craniotomy was established before the development of improved bipolar and hemostatic agents in use today.46 Additionally, image guidance and more specifically diffusion tractography make it possible to identify landmarks with greater certainty. Technology such as the Vycor (Vycor Medical Inc., Boca Raton, Florida) retractor system and the BrainPath (NICO Corporation, Indianapolis, Indiana) tube retractor system are utilized at our institution for small, noninfiltrative lesions below the cortical surface. We consider this a separate approach, appropriate for other types of pathology. Our approach to glioma surgery is to attempt supramaximal resection, which in most cases obviates the need for tubular and/or hinged blade retractor systems given that the cortical surface, when involved, is included in the resection. A longer operating plane is also afforded through our technique. Limitations One of the limitations of our technique is its dependence on a surgeon's comfort with operating through a small craniotomy. Keyhole exposure is a dynamic process that requires operative preparation prior to undertaking surgery in real patients for most surgeons, particularly with respect to what Teo refers to as “anticipatory positioning” with the operative microscope.13 This requires careful planning and involves a small learning curve. Other limitations include the use of diffusion tractography with surgical image guidance, which is subject to error. Additionally, in patients with previous craniotomies not done utilizing this technique, incision planning can be quite challenging in order to avoid scalp necrosis. We note again that this study aims to report the plausibility of resecting gliomas through a frontal keyhole craniotomy with good outcomes, and does not provide evidence that this method is superior to traditional craniotomy. Sufficient data to support this claim in a single series will prove difficult to attain. Additionally, frontal keyhole surgery is only of benefit in carefully selected patients and in the hands of an experienced surgeon. We present our results as we feel this technique is both safe and effective for treating frontal gliomas. CONCLUSION We present evidence that low- and high-grade gliomas can be maximally resected implementing the frontal keyhole approach without additional risk to the patient. Neurosurgeons can achieve outcomes similar to those with larger, traditional craniotomies in some patients presenting with frontal gliomas. Disclosure The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article. REFERENCES 1. Marcus HJ, Cundy TP, Hughes-Hallett A, Yang GZ, Darzi A, Nandi D. Endoscopic and keyhole endoscope-assisted neurosurgical approaches: a qualitative survey on technical challenges and technological solutions. Br J Neurosurg . 2014; 28( 5): 606- 610. Google Scholar CrossRef Search ADS PubMed 2. Duffau H. A new philosophy in surgery for diffuse low-grade glioma (DLGG): oncological and functional outcomes. Neurochirurgie . 2013; 59( 1): 2- 8. Google Scholar CrossRef Search ADS PubMed 3. Duffau H. A new concept of diffuse (low-grade) glioma surgery. Adv Tech Stand Neurosurg . 2012; 38: 3- 27. Google Scholar CrossRef Search ADS PubMed 4. Bonney PA, Conner AK, Boettcher LB, Pittman, N, Sughrue M. A simplified method of accurate postprocessing of diffusion tensor imaging for use in Brain tumor resection. Oper Neurosurg . 2015; 17( suppl 5). 5. Oya S, Kim SH, Sade B, Lee JH. The natural history of intracranial meningiomas. J Neurosurg . 2011; 114( 5): 1250- 1256. Google Scholar CrossRef Search ADS PubMed 6. Boling W. Minimal access keyhole surgery for mesial temporal lobe epilepsy. J Clin Neurosci . 2010; 17( 9): 1180- 1184. Google Scholar CrossRef Search ADS PubMed 7. Bonney PA, Boettcher LB, Cheema AA, Maurer AJ, Sughrue ME. Operative results of keyhole supracerebellar-infratentorial approach to the pineal region. J Clin Neurosci . 2015; 22( 7): 1105- 1110. Google Scholar CrossRef Search ADS PubMed 8. Cheng CM, Noguchi A, Dogan A et al. Quantitative verification of the keyhole concept: a comparison of area of exposure in the parasellar region via supraorbital keyhole, frontotemporal pterional, and supraorbital approaches. J Neurosurg . 2013; 118( 2): 264- 269. Google Scholar CrossRef Search ADS PubMed 9. Renovanz M, Hickmann AK, Gutenberg A, Bittl M, Hopf NJ. Does size matter? Minimally invasive approach in pediatric neurosurgery–a review of 125 minimally invasive surgeries in children: clinical history and operative results. Childs Nerv Syst . 2015; 31( 5): 665- 674. Google Scholar CrossRef Search ADS PubMed 10. Reisch R, Perneczky A. Ten-year experience with the supraorbital subfrontal approach through an eyebrow skin incision. Neurosurgery . 2005; 57( 4 suppl): 242- 255; discussion 242-255. Google Scholar PubMed 11. Tomasello F, Cardali S, Angileri FF. The quest of minimally invasive neurosurgery. World Neurosurg . 2014; 81( 2): 263- 264. Google Scholar CrossRef Search ADS PubMed 12. Hopf NJ, Reisch R. Axel Perneczky, 1.11.1945-24.1.2009. Minim Invasive Neurosurg . 2009; 52( 1): 1- 4. Google Scholar CrossRef Search ADS PubMed 13. Teo C, Sughrue, ME. Principles and Practice of Keyhole Brain Surgery . Stuttgart: Thieme; 2015. Google Scholar CrossRef Search ADS 14. Sughrue ME, Bonney PA, Choi L, Teo C. Early discharge after surgery for intra-axial brain tumors. World Neurosurg . 2015; 84( 2): 505- 510. Google Scholar CrossRef Search ADS PubMed 15. Clark JC, Spetzler RF. Defining the limits of the occipital transtentorial keyhole approach. World Neurosurg . 2013; 80( 1-2): 62- 63. Google Scholar CrossRef Search ADS PubMed 16. Daming C, Yiwen S, Bin Z et al. Large vestibular schwannoma resection through the suboccipital retrosigmoid keyhole approach. J Craniofac Surg . 2014; 25( 2): 463- 468. Google Scholar CrossRef Search ADS PubMed 17. Kang HJ, Lee YS, Suh SJ, Lee JH, Ryu KY, Kang DG. Comparative analysis of the mini-pterional and supraorbital keyhole craniotomies for unruptured aneurysms with numeric measurements of their geometric configurations. J Cerebrovasc Endovasc Neurosurg . 2013; 15( 1): 5- 12. Google Scholar CrossRef Search ADS PubMed 18. Ma Y, Lan Q. An anatomic study of the occipital transtentorial keyhole approach. World Neurosurg . 2013; 80( 1-2): 183- 189. Google Scholar CrossRef Search ADS PubMed 19. Marcus HJ, Sarkar H, Mindermann T, Reisch R. Keyhole supracerebellar transtentorial transcollateral sulcus approach to the lateral ventricle. Neurosurgery . 2013; 73( 2 Suppl Operative): onsE295- onsE301; discussion onsE301. Google Scholar PubMed 20. Moscovici S, Mizrahi CJ, Margolin E, Spektor S. Modified pterional craniotomy without "MacCarty keyhole". J Clin Neurosci . 2016; 24: 135- 137. Google Scholar CrossRef Search ADS PubMed 21. Ditzel Filho LF, McLaughlin N, Bresson D, Solari D, Kassam AB, Kelly DF. Supraorbital eyebrow craniotomy for removal of intraaxial frontal brain tumors: a technical note. World Neurosurg . 2014; 81( 2): 348- 356. Google Scholar CrossRef Search ADS PubMed 22. Fischer G, Stadie A, Reisch R et al. The keyhole concept in aneurysm surgery: results of the past 20 years. Neurosurgery . 2011; 68( 1 Suppl Operative): 45- 51; discussion 51. Google Scholar PubMed 23. Maurer AJ, Bonney PA, Strickland AE, Safavi-Abbasi S, Sughrue ME. Brainstem cavernous malformations resected via miniature craniotomies: technique and approach selection. J Clin Neurosci . 2015; 22( 5): 865- 871. Google Scholar CrossRef Search ADS PubMed 24. Reisch R, Fischer G, Stadie A, Kockro R, Cesnulis E, Hopf N. The supraorbital endoscopic approach for aneurysms. World Neurosurg . 2014; 82( 6 suppl): S130- S137. Google Scholar CrossRef Search ADS PubMed 25. Wilson DA, Duong H, Teo C, Kelly DF. The supraorbital endoscopic approach for tumors. World Neurosurg . 2014; 82( 6 suppl): S72- S80. Google Scholar CrossRef Search ADS PubMed 26. Yu LH, Yao PS, Zheng SF, Kang DZ. Retractorless surgery for anterior circulation aneurysms via a pterional keyhole approach. World Neurosurg . 2015; 84( 6): 1779- 1784. Google Scholar CrossRef Search ADS PubMed 27. Aghi MK, Nahed BV, Sloan AE, Ryken TC, Kalkanis SN, Olson JJ. The role of surgery in the management of patients with diffuse low grade glioma: a systematic review and evidence-based clinical practice guideline. J Neurooncol . 2015; 125( 3): 503- 530. Google Scholar CrossRef Search ADS PubMed 28. Hervey-Jumper SL, Berger MS. Role of surgical resection in low- and high-grade gliomas. Curr Treat Options Neurol . 2014; 16( 4): 284. Google Scholar CrossRef Search ADS PubMed 29. Watts C. Surgical management of high-grade glioma: a standard of care. CNS Oncol . 2012; 1( 2): 181- 192. Google Scholar CrossRef Search ADS PubMed 30. Hervey-Jumper SL, Berger MS. Reoperation for recurrent high-grade glioma: a current perspective of the literature. Neurosurgery . 2014; 75( 5): 491- 499; discussion 498-499. Google Scholar CrossRef Search ADS PubMed 31. Sughrue ME, Sheean T, Bonney PA, Maurer AJ, Teo C. Aggressive repeat surgery for focally recurrent primary glioblastoma: outcomes and theoretical framework. Neurosurg Focus . 2015; 38( 3): E11. Google Scholar CrossRef Search ADS PubMed 32. Beseoglu K, Lodes S, Stummer W, Steiger HJ, Hanggi D. The transorbital keyhole approach: early and long-term outcome analysis of approach-related morbidity and cosmetic results. Technical note. J Neurosurg . 2011; 114( 3): 852- 856. Google Scholar CrossRef Search ADS PubMed 33. Burks JD, Bonney PA, Conner AK et al. A method for safely resecting anterior butterfly gliomas: the surgical anatomy of the default mode network and the relevance of its preservation. J Neurosurg . 2016; 1- 17. 34. Hervey-Jumper SL, Li J, Lau D et al. Awake craniotomy to maximize glioma resection: methods and technical nuances over a 27-year period. J Neurosurg . 2015; 123( 2): 325- 339. Google Scholar CrossRef Search ADS PubMed 35. Sanai N, Chang S, Berger MS. Low-grade gliomas in adults. J Neurosurg . 2011; 115( 5): 948- 965. Google Scholar CrossRef Search ADS PubMed 36. Sanai N, Berger MS. Extent of resection influences outcomes for patients with gliomas. Rev Neurol (Paris) . 2011; 167( 10): 648- 654. Google Scholar CrossRef Search ADS PubMed 37. Sanai N, Berger MS. Glioma extent of resection and its impact on patient outcome. Neurosurgery . 2008; 62( 4): 753- 764; discussion 264-756. Google Scholar CrossRef Search ADS PubMed 38. Abolfotoh M, Horowitz PM, Chiocca EA. Awake craniotomy and intraoperative mri for maximal safe resection in a case of an extensive left frontal and insular low-grade glioma: 3-dimensional operative video. Neurosurgery . 2015; 11( 4). 39. Mohammadi AM, Sullivan TB, Barnett GH et al. Use of high-field intraoperative magnetic resonance imaging to enhance the extent of resection of enhancing and nonenhancing gliomas. Neurosurgery . 2014; 74( 4): 339- 348; discussion 349; quiz 349-350. Google Scholar CrossRef Search ADS PubMed 40. Olubiyi OI, Ozdemir A, Incekara F et al. Intraoperative magnetic resonance imaging in intracranial glioma resection: a single-center, retrospective blinded volumetric study. World Neurosurg . 2015; 84( 2): 528- 536. Google Scholar CrossRef Search ADS PubMed 41. Paiva WS, Fonoff ET, Marcolin MA, Cabrera HN, Teixeira MJ. Cortical mapping with navigated transcranial magnetic stimulation in low-grade glioma surgery. Neuropsychiatr Dis Treat . 2012; 8: 197- 201. Google Scholar CrossRef Search ADS PubMed 42. Pamir MN, Ozduman K, Yildiz E, Sav A, Dincer A. Intraoperative magnetic resonance spectroscopy for identification of residual tumor during low-grade glioma surgery: clinical article. J Neurosurg . 2013; 118( 6): 1191- 1198. Google Scholar CrossRef Search ADS PubMed 43. Ius T, Isola M, Budai R et al. Low-grade glioma surgery in eloquent areas: volumetric analysis of extent of resection and its impact on overall survival. A single-institution experience in 190 patients: clinical article. J Neurosurg . 2012; 117( 6): 1039- 1052. Google Scholar CrossRef Search ADS PubMed 44. Odland A, Server A, Saxhaug C et al. Volumetric glioma quantification: comparison of manual and semi-automatic tumor segmentation for the quantification of tumor growth. Acta Radiol . 2015; 56( 11): 1396- 1403. Google Scholar CrossRef Search ADS PubMed 45. Eyupoglu IY, Hore N, Merkel A, Buslei R, Buchfelder M, Savaskan N. Supra-complete surgery via dual intraoperative visualization approach (DiVA) prolongs patient survival in glioblastoma. Oncotarget . 2016; 7( 18): 25755- 25768. Google Scholar CrossRef Search ADS PubMed 46. Teo C. The concept of minimally invasive neurosurgery. Neurosurg Clin N Am . 2010; 21( 4): 583- 584. Google Scholar CrossRef Search ADS PubMed Copyright © 2017 by the Congress of Neurological Surgeons
Neurosurgery – Oxford University Press
Published: Mar 1, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera