Is “En Masse” Tumor Resection a Safe Surgical Strategy for Low-Grade Gliomas? Feasibility Report on 74 Patients Treated Over Four Years
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.314527
Source of Support: None, Conflict of Interest: None
Keywords: Low-grade glioma, motor cortex, euronavigation, surgery, white fiber tractsKey Message: Gliomas arise from white matter tracts and grow in an expansile fashion that displaces the adjoining tracts. The color, consistency, and vascularity of these tumors permit the identification of a plane of dissection and a possibility of en masse tumor resection.
Based on our recent understanding, we speculate that gliomas are “confined” tumors. They arise from white matter tracts and grow in an expansile and “disciplined” fashion along the involved tract and gyrus while displacing the adjacent tracts around it.,,,, The gliomas tend to be “localized” when they arise from short arcuate fibers and remain confined to the affected gyrus/gyri. Gliomas arising from long association and commissural fibers follow a similar pattern of growth and expansion but are more “diffuse” as they spread along the length of the tract in the hemisphere. When the tumor involves the commissural fibers the glioma propagation tends to be bilateral. Both low- and high-grade gliomas have a similar pattern of expansion and well-defined boundaries and are different only in their growth pattern. The surgical strategy was based on the concept that gliomas arise from and extend along a named white fiber tract and grow by circumferential expansion. The adjoining fiber tracts are displaced, limit and “contain” the spread of the tumor and are functionally involved by compression and deformation and not by invasion, involvement, or destruction.
Herein, we report our surgical strategy for the treatment of low-grade gliomas involving the short arcuate white fibers. Our literature search did not reveal any report describing a similar surgical strategy for low-grade gliomas.
During the period from January 2016 to June 2019, 74 patients harboring “low-grade” gliomas (i.e., a measurable fluid-attenuated inversion recovery [FLAIR] mass lesion without any postcontrast enhancement) that involved the short arcuate fibers in various lobes of the cerebral hemispheres were surgically treated in the neurosurgical departments of the authors. Tumors involving the insula and “diffuse” group of tumors arising from the long association fibers and the commissural fibers were analyzed in a separate study and have not been included here. World Health Organization (WHO) Grade I tumors such as pilocytic astrocytoma and “high-grade” gliomas were excluded from the analysis. [Table 1] identifies the location of the gyrus harboring the tumor-affected short arcuate fibers.
Mini-mental state examination (MMSE) was used to assess neurological and cognitive status. The patients were scored in the immediate postoperative period and at every follow-up.
Preoperative investigations included a plain, FLAIR, and gadolinium-enhanced magnetic resonance imaging (MRI) with MR spectroscopy, MR perfusion evaluation, and diffusion tensor imaging. Functional MRI was performed in 24 patients with tumors in eloquent locations. Tractography was performed in all patients on the Brainlab planning software. The tumors were classified according to their fiber tract of origin and based on the gyrus involved.
Adjuncts during surgery
Neuronavigation and Tractography: Neuronavigation was performed in all patients using a Brainlab Vector Vision planning station. Tractography was performed using the iPlan Fiber Tracking software. By using several seed points located in one or more regions of interest (ROIs), the connecting fiber bundles were tracked using fiber assignment by continuous tracking (FACT) approach.
Ultrasound: Real-time ultrasound monitoring was done using a Samsung real-time portable scanner before handling the brain and after the completion of the surgical tumor resection in all patients.
Intraoperative neurophysiology: After induction of total intravenous anesthesia, corkscrew electrodes were applied over the ipsilateral motor area and subdermal electrodes were placed on the contralateral face, upper, and lower limbs. Motor mapping was performed in 14 cases and was done mainly for tumors in the vicinity of the motor cortex. Multimodality neurophysiologic monitoring was done using direct and subcortical monopolar and bipolar stimulation, transcranial motor evoked potentials (MEPs), and somatosensory evoked potentials (SSEPs) (NIM - Eclipse, Medtronic, Dublin, Ireland). High-frequency transcranial MEPs were measured at intervals of 3–5 min during the tumor resection using a stimulation intensity of 300 to 500 volts. Bipolar cortical mapping was performed using a 5 mm bipolar probe using a biphasic square wave pulse of 1 ms duration at 60 Hz. Stimulation was begun at 4 mA and increased to a maximum of 16 mA sequentially. Subcortical mapping with the monopolar probe was performed in seven patients. Awake craniotomy for language testing was performed in 11 patients, wherein the tumor was in proximity to the motor speech areas (7 patients) and the vicinity of the arcuate fasciculus (1 patient) or the inferior fronto-occipital fasciculus (3 patients) on the dominant side. Stimulation was performed with 60 Hz using a bipolar probe starting at 1 mA and then increasing up to 6 mA in 0.5 mA increments. The stimulation was performed on the cortex at 10 mm intervals. Intraoperative fluorescence to evaluate the extent of tumor resection was not used in any case. Intraoperative MRI was not available.
The surgical strategy essentially involved the following three steps: 1) To identify the gyrus/gyri associated with the tumor affected short arcuate fibers, 2) Understanding the location and displacement of long association, projection, and commissural fibers in the vicinity of the tumor, and 3) Performing an en masse and radical surgical resection by developing a plane of dissection between the tumor and adjoining normal brain and keeping in perspective the subcortical anatomical boundaries formed by the displaced tracts and cortical eloquence. Although the methodology of surgical procedure was essentially individual case-based and cannot be generalized, the surgical principle was common. The location of the tumor was identified by its color, consistency, and vascularity. Intraoperative ultrasound and navigation tools were used to confirm the location and extent of the tumor. Low-grade gliomas generally appear “whiter” and were visually distinguishable from normal adjoining normal brain. The tumors were firmer and gritty in consistency and were relatively avascular. A plane of dissection of the tumor from the adjoining brain was identified and developed using blunt surgical dissection. The sulcal blood vessels were displaced by the tumor and could be protected. Experience in the identification of normal from abnormal and developing the dissection plane appeared key surgical steps. The dissection plane between tumor and normal brain was reliable and safe and allowed radical tumor resection. For tumors that were in proximity to the “functional” cortical motor area, a tailored customized resection was performed whenever there was even moderate difficulty in identifying the plane of dissection or when there were negative indicators by neurophysiological monitoring.
There were 46 males and 28 females. The ages of the patients ranged from 21 to 55 years (average 33 years). The duration of symptoms ranged from ten days to five years. [Table 2] summarizes the clinical features at the time of presentation. The location of the tumors, the fiber tracts in the vicinity, and the nature of tumor resection are summarized in [Table 1]. Visual impression and consistency of tumor allowed the development of a plane and radical tumor resection. The strategy of en masse resection entailed the removal of the entire affected gyrus in the region of the involved short arcuate fibers. Essentially, this surgical strategy of developing a surgical plane between normal brain and glioma led to a “supramarginal” resection in 25 patients.
Extent of resection
Postoperative MRI was performed in all patients within 48 h of surgery and again at an interval of three months to gauge the extent of resection [Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5]. The preoperative and postoperative FLAIR images were compared to quantify the resection. The resection was labeled as subtotal (more than 15 mL of tumor residue) when some portion of the tumor was left behind to preserve function or was left behind inadvertently.,, A supramaximal resection was defined as resection of all signal abnormality and when the immediate postoperative resection cavity was bigger than the preoperative tumor cavity. A gross total resection was defined as resection of all signal abnormality and when the tumor resection cavity was the same as the preoperative tumor volume. An essentially en masse tumor resection was possible in 47 cases. Twelve patients had a subtotal resection. Thirty-seven patients had gross total tumor resection and 25 patients had a supramaximal resection [Table 3].
The follow-up ranged from 11 to 56 months (mean 21 months). All patients improved in their preoperative complaints of headache and seizures following surgery. Three patients with motor weakness improved in their neurological deficits following surgery. None of the patients had any decline in their cognitive functions and were able to lead fully functional lives.
Out of the three patients who had preoperative speech deficit, in two patients the tumor was located in the left inferior frontal gyrus, and in one patient the tumor was in the fusiform gyrus. All these three patients improved in their speech function following surgery. One patient with a lesion in the pars opercularis and without any preoperative speech deficit developed anarthria with stimulation during awake surgery and underwent subtotal tumor resection. Immediately following surgery, the patient had dysarthria, which improved at follow-up and the patient had a normal speech at follow-up of two years. Two patients with tumors located in the posterior superior frontal gyrus developed supplementary motor area (SMA) syndrome after surgery but recovered completely within three months. Postoperative MRI performed in the patients with the new deficits did not reveal the presence of any ischemic injury or hemorrhage.
Postoperative MRI scans were performed 48 h after surgery, at three months, and six monthly intervals thereafter. Patients with complete tumor excision on FLAIR imaging were clinically observed without any use of adjuvant treatment. The histology of the patients was WHO Grade II astrocytoma (33 patients) or oligodendroglioma (41 patients). Twelve patients who had undergone a subtotal resection were given adjuvant radiotherapy and chemotherapy. At a follow-up of 11–52 months (average 30 months), the tumor residue was found to be stable in this group of patients. One patient with a medial temporal tumor developed a recurrence of the tumor 26 months after the first surgery and was reoperated. The histology was Grade II diffuse astrocytoma. Postoperative MRI showed a total resection of the recurrent tumor. None of the other patients had any recurrence or growth of the residual tumor.
The general understanding is that low-grade gliomas are by nature “diffuse,” “invasive,” or “infiltrating” into the normal brain parenchyma. Considering their frequently identified wide extension and the location in critical, functional, or eloquent brain areas, injury to subcortical pathways can be responsible for morbidity encountered during surgical resection.,,,,,,,,,,,,,,,, Safe resection of tumors near functional areas necessarily involves intraoperative identification of functional sites both at the cortical and the subcortical levels. For several years, the generally agreed-upon or gold standard treatment strategy for low-grade glioma was either symptomatic drug treatment and clinical observation or a biopsy for histological confirmation that was then followed up with adjuvant treatment. The current view is that radical resection of low-grade gliomas is “safely” possible and is associated with recovery from symptoms, improved longevity, and quality of life, and decreased potential for “malignant” transformation. Moreover, adjuvant chemotherapy/radiotherapy treatment can be avoided. Technical adjuncts such as awake surgery and use of 5-ALA, real-time ultrasound, cortical and subcortical mapping, and intraoperative MRI are currently in vogue to achieve the goal of maximal tumor resection with minimum or no morbidity.
Based on an anatomical understanding of the three-dimensional architecture of the white fibers of the brain and our surgical experience in dealing with low-grade gliomas we speculate that the point of origin of glioma is from the white fiber tracts of the brain. The propagation of the glioma occurs in the direction of the involved fiber bundle.
From our study of gliomas over the years, we realized that like the benign brain tumors viz. meningiomas and neurinomas, gliomas grow by expansion into the adjoining gyri and displace the adjoining fiber tracts and provide a well-defined plane of surgical dissection around its circumference. This concept is in variance with the generally agreed viewpoint that gliomas merge into the adjoining normal brain parenchyma and encase normal blood vessels.
Surgery was essentially based on the understanding of the anatomical framework of white fiber tracts. By using the anatomical fiber tract model and information obtained from tractography and functional MRI, it was possible to perform en masse tumor resection in 47 (63.5%) patients with “minimal” or no neurological morbidity. The resection was supramarginal in 25 (33.8%) patients. In a recent study, Rossi et al. assessed the feasibility of supratotal resection in low-grade gliomas. The location of the tumor was the most important factor in determining the feasibility of “en masse” supratotal resection in our series. We could achieve an “en masse” total or supratotal resection in all the patients with medial and basal frontal tumors, in basal temporal tumors, and medial parietal and occipital tumors. Tumors involving or in the vicinity of the eloquent cortex were operated with the same surgical strategy of en masse resection with more liberal use of neurophysiological monitoring. Awake surgery was used for patients with tumors in the left hemisphere in the region of the inferior frontal gyrus, superior temporal gyrus, and angular gyrus.
Identification of normal from abnormal, dissection around the surface of the tumor, preservation of blood vessels coursing in the vicinity are important prerequisites of successful en masse resection. The plane of dissection between a normal brain and an abnormal tumor-involved brain was identified to be well defined and reliable. In all the patient's tractography showed that the long association fibers, the commissural fibers, and the projection fibers were essentially displaced by the tumor.
The sulcul blood vessels are located around the dome of the tumor and encased within the tumor lobulations and need to be preserved as they are not “tumor vessels” but are normal coursing, en passage, or transit blood vessels. While neurological deficits related to brain handling and manipulation usually recover over time, deficits due to a vascular injury are generally long-standing.
Improvement in neurological symptoms (95.9% patients) following surgical tumor resection is indicative of the fact that the displaced tracts were functionally compromised due to pressure and not due to destruction, invasion, or involvement. There was an improvement in motor deficits and speech in six patients. This may even be attributed to the normalization, reorganization, and plasticity of functional cortical areas around the dome of the tumor. None of our patients had any recognizable deterioration in cognitive and psychological function and were able to regain active and functional lives. This suggests that the tumor-affected brain was essentially “nonfunctional.” Histological examination of the resected tumor specimen at multiple areas within the tumor bulk showed a tumor in all components and the absence of normal brain configuration. Though postoperative tractography can show the realignment of the deformed tracts, following surgery it was not performed in any of our patients during the study period. However, more recently this modality of imaging is used in most patients to analyze the status and realignment of the tracts.
Intraoperative navigation and real-time ultrasound formed important tools to identify the location, extent, and depth of the tumor. Although these technological tools were useful and were used, the color and consistency of the tumor were identified to be more reliable to locate the tumor. As the experience with en masse tumor resection matured and the dissection technique was perfected, low-grade gliomas even in eloquent motor and speech areas were resected using the same technique. Six cases of gliomas located in the premotor area, both on the left and right side, were resected by en masse technique with the assistance of neuromonitoring. In none of the cases was there any new motor neurological deficit following surgery. Twelve patients underwent subtotal resection. In six out of these 12 patients, the tumor was left behind in eloquent regions based on the results of brain mapping. Although an attempt was made to resect the tumor radically and by en masse technique, postoperative imaging showed the presence of residual tumor in six cases. This suggests that either the tumor extended into adjoining parenchyma with tail-like protrusions or the plane of surgical dissection was lost during surgery. From our initial experience, it appears that the fear of affecting neural damage to critical adjoining tracts, a conservative surgical resection was probably carried out and the plane of dissection was lost at some points.
Some authors have advocated a subpial en bloc – sulcal-to-sulcal resection when possible as it maintains margins throughout the tumor removal without interfering with tumor vascularity. In tumors near the eloquent cortex, a piecemeal resection is advocated.
Although intraoperative neurophysiological monitoring and imaging were used in the presented cases, it appears that with more experience in dissecting normal from abnormal and by the technique of en masse resection, both these intraoperative procedures are dispensable. It also appears that with more experience in developing a plane between the tumor and adjoining normal structures, the need for awake craniotomy can be minimized or avoided.
The technique of “en masse” resection of low-grade glioma is possible, safe, and ensures radical tumor resection. Real-time anatomical information of the involved and displaced white fiber tracts is crucial for adopting such a surgical strategy.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5]
[Table 1], [Table 2], [Table 3]