The concept of the “Hexahedron” in the supratotal resection of the frontal gliomas: A technical note

OBJECTIVE To evaluate the value of the concept of the “Hexahedron” in the supratotal resection (SPTR) of frontal gliomas in both dominant and nondominant hemispheres . METHODS All consecutive patients who underwent SPTR for frontal gliomas under the guidance from the concept of the “Hexahedron” were retrospectively analysed for lesion location, pathology, extent of resection (EOR), and complications from May 2020 to June 2021. Volumetric EOR was measured and classied as SPTR, (in which the volume of the postoperative cavity was larger than the preoperative tumour volume), gross total resection (GTR, > 95% by volume) or subtotal resection (STR, ≤ 95% by volume) after independent radiological review.


Introduction
The management of gliomas has been extensively studied, and consistently, surgery is said to play a central role in their treatment. The consistency of available data supports the long-held principle that safe, maximal tumour resection including high-grade (HGG) and low-grade glioma (LGG) improves symptom management, quality of life, progression-free survival (PFS), and overall survival (OS) in glioma patients [1].
For maximally resecting frontal gliomas, we proposed the concept of the "Hexahedron" based on their growth range and the functional characteristics of the frontal lobe. The six sides and different boundaries of the frontal resection area were de ned following guidance from the "Hexahedron" concept.

Material And Methods
This retrospective study included all consecutive patients who underwent the resection of the frontal gliomas under guidance from the concept of the "Hexahedron" at Chongqing Sanbo Jiangling Hospital and Yunnan Cancer Hospital, China between May 2020 and June 2021. There were no age exclusion criteria. A total of eight patients were enrolled in this study, including six males and two females, with an average age of 47.13 (26 69) years. Patient medical records, clinical visits, and imaging studies were reviewed until the last available follow-up. Tumour pathology, postoperative complications, and clinical outcomes were collected (Table 1). Resection rates were considered as follows: supra-total removal (SPTR), in which the margin for tumour resection is extended beyond the abnormal area detected by magnetic resonance imaging (MRI); gross total removal (GTR), where no residual tumour is visible on immediate postoperative MRI; and subtotal removal (STR), where < 90% of the lesion is resected. This retrospective study was approved by the Academic Committee of Chongqing Sanbo Jiangling Hospital.

Surgical Technique
With the patient in the supine position, a coronary incision was performed, and the ap was reversed forward. The frontal and temporal bone windows were exposed, 2 holes were drilled in the frontal bone, and a bone ap measuring approximately 10x7cm 2 was removed with a bone knife. The lower boundary of the bone ap was near the superior part of the frontal sinus, which was not left open. The superior boundary of the bone ap was close to the coronal suture, the medial boundary was attached to the midline, and the lateral boundary was attached to the squamous suture.
The dura was opened and turned to the midline and anterior skull base to expose the frontal lobe and lateral ssure. The differences between normal brain tissue and tumour tissue were identi ed carefully under a microscope. Some tumours could be identi ed by their appearance: LGGs were greyish white, while HGGs were greyish red. Some tumours were not easy to distinguish visually from normal brain tissue and thus had to be identi ed by differences in texture and blood supply and preoperative anatomical localization.
Tumors with clear boundaries were removed along the tumour boundary. First, the medial side of the tumour was removed as most of the blood supply of frontal gliomas are derived from the pericallosal artery and the callosomarginal artery; removing the medial side of the tumour was thus equivalent to disconnecting its main blood supply. Then, the boundary between the tumour and brain tissue was gradually pushed apart with cotton strips or gelatine sponge, allowing removal of the entire tumour. If the tumour partially crossed the midline, however, it was di cult to remove the tumour in one piece on the opposite side of the midline, and it thus had to be resected in blocks. In these cases, the branches of the pericallosal artery and the callosomarginal artery needed to be carefully identi ed to protect large branches ( Fig. 1) (Fig. 2). Some tumours invaded the basal ganglia and the thalamus, and were thus also di cult to remove.
After the tumour was resected, extended resection of the surrounding frontal lobe tissue was performed.
The whole resection range could be de ned as an irregular hexahedron ( For gliomas with unclear boundaries and in ltration to the lateral ventricle, basal ganglia, thalamus and the opposite side of the midline, complete removal while maintain the integrity of important functions was impossible; this also applied to tumours that had partially in ltrated the basal ganglia and the thalamus. In these cases, our overall surgical strategy was to remove most of the tumour tissue and the resectable portion of the frontal lobe within a safe range according to the concept of the "Hexahedron". During the operation, the actual surgical strategy was to remove part of the tumour tissue in blocks that were not along its boundary. Then, under guidance from the concept of the Hexahedron, the tissue of the frontal lobes were removed to the greatest extent possible in a safe range along the six surfaces.

Results
The pathological anatomical diagnosis was glioblastoma WHO Grade IV in 4 patients, anaplastic oligodendroglioma WHO Grade III in 1, anaplastic astrocytoma WHO Grade III in 2 and diffuse astrocytoma WHO Grade II in 1.
For postoperative imaging, high-grade gliomas were mainly analysed with enhanced MRI, and low-grade gliomas were analysed with T2 uid-attenuated inversion recovery (FLAIR) sequences. Achievement of SPTR was de ned as tumour resection extending beyond the abnormal MRI-veri ed area, which indicated that the volume of the postoperative cavity was larger than the preoperative tumour volume. Six patients received SPTR and two patients received STR (due to invasion of the basal ganglia and thalamus). In the two patients who underwent STR, except for the part invading the basal ganglia and the thalamus, the other interfaces of the tumour were resected according to the concept of the Hexahedron.
The postoperative imaging of the six patients who received SPTR was basically consistent with the range of the Hexahedron that we determined during the operation.
The main postoperative complications were contralateral paresis in two patients and memory disturbances in one patient. There were no cases of rebleeding and secondary operation during hospitalization.

Adjuvant Therapies
In our study, postoperative pathology revealed diffuse astrocytoma WHO Grade II in one patient, which received STR. The other seven patients had WHO Grade III and Grade IV tumors. These seven patients underwent adjuvant radiotherapy and chemotherapy after the operation, whlie the eighth patient underwent adjuvant radiotherapy alone. The patient who achieved STR received radiotherapy, chemotherapy and tumour treating elds (TTF) therapy.
Long-term Follow-up Surgical resection is now considered for asymptomatic, incidentally found low-grade gliomas. A few studies favour increasing the extent of resection to improve overall and progression-free survival in cases of low-grade glioma [2][3][4][5][6][7][8][9][10]. These studies showed a mean survival bene t from 61.1 to 90 months with maximal resection. A large population-based natural history study of Norwegian patients offered additional insight in their analysis of 153 glioma patients treated in two hospitals serving adjacent geographical regions [4,11], where the approach to the care of gliomas in individualized patients was dependent on residential address. The median survival in this study was 5.9 years for patients receiving tumour biopsy, while the group receiving early resection did not reach median survival by the end of the study period, suggesting a survival bene t for those treated with early surgery. Furthermore, the 5-year survival was 60% for biopsy patients and 74% for those receiving early surgery. GTR impacts the natural history of low grade glioma. The time to malignant transformation for low-grade gliomas ranges from 4 to 29 months, and approximately 45% of patients with diffuse low-grade WHO grade II glioma will We reviewed the signi cance of the extent of resection in improving overall and progression-free survival for patients with primary and recurrent high-grade gliomas found in a few reports [1,[14][15][16][17][18][19]. The preponderance of evidence supports using the extent of resection as a signi cant predictor of overall and progression-free survival in high-grade glioma. After GTR, overall survival improves to 64.9-75.2 months for WHO grade III gliomas and to 11.3-18.5 months for WHO grade IV gliomas [20,21]. Although limited by selection bias, a retrospective analysis of 107 patients with recurrent glioblastoma revealed that patients who initially received subtotal resection experienced a survival bene t when GTR was achieved at recurrence (19 months overall survival for GTR vs. 15.9 months overall survival for STR) [22]. It has been further suggested that an extent of resection threshold of at least 80% is the minimum needed to offer a survival bene t in recurrent glioblastoma [23].
GTR or SPTR can be di cult to achieve in many cases due to glioma in ltration into cortical and subcortical regions important for language, motor, and neurocognitive function. However, neurosurgeons should strive to improve the surgical technology and achieve GTR or SPTR of glioma. A rm understanding of neuroanatomy is critical to performing a safe operation for glioma resection.
Our research focused on the frontal lobe which are the most common site of gliomas. GTR can be performed for frontal gliomas with clear boundaries, while for gliomas growing in the direction of the paraventricular nucleus, basal ganglia or thalamus, complete removal of the parts in ltrating into the deep structures can be di cult. Existing relevant studies have re ected the idea of extended glioma resection [1,3,5,16,18,24], which can reduce the recurrence rate of gliomas and prolong the survival time of the patients. Due to the relatively wide space in the frontal lobe, it is possible to expand the resection of frontal gliomas. Based on the above ideas, removing the frontal tumour and frontal lobe tissue as much as possible within the safe range without damaging the important functions of movement and language can be challenging.
According to our previous surgical experience, we de ned the maximum range of safe resection for frontal gliomas from an anatomical perspective: the anterior skull base as the anterior boundary, the midline as the medial boundary, the inferior frontal sulcus as the lateral boundary, and the coronal suture as the posterior boundary; resection according to these safe boundaries cannot cause damage to important functions. Based on this idea, we propose the concept of the Hexahedron for maximum safe resection of frontal gliomas. The Hexahedron is an irregular hexahedron surrounded by 6 surfaces and 8 points. The anterior surface (frontal sinus surface) is surrounded by the nasal frontal suture, the lateral orbital point, the medial point above the frontal sinus and the root of the sphenoid ridge. The front range is the smallest surface, and includes the front end of the straight gyrus and the orbital gyrus. The location of the frontal surface is deep and does not need to be completely exposed during craniotomy. This part can be removed by adjusting the angle of the microscope during operation. The inferior surface On preoperative imaging, some tumours have clear approximate boundaries, which can be completely moved under the microscope. The resection strategy for this kind of tumour is to remove it along the tumour boundary rst, which is helpful for comlete tumour removal and avoiding residual tumour. Then, the resection is expanded to the six surfaces according to the range of the Hexahedron.
For tumours diffusely in ltrating the ventricles, basal ganglia, thalamus and contralateral hemisphere, a part of the tumours is removed rst, and then most of the remaining tumours and the expandable frontal lobe are removed together according to the guidance of the Hexahedron.
Many skills are involved in the process of tumour resection, and the experience in different neurosurgical centres is different. According to our experience, the operators should rst focus on removing the straight gyrus and orbital gyrus piece by piece at the bottom of the frontal lobe under the arachnoid when close to the inferior surface of the Hexahedron (sphenoid plateau). It is best to leave a thin layer of arachnoid to observe and avoid damaging the deep optic nerve. When close to the posterior surface of the Hexahedron (hypothalamus and ventricular surface), the position of the bottom of the hypothalamus and the front of the hypothalamus can be determined by the direction of the optic nerve. By removing the thin layer of tissue in front of the frontal horn of the lateral ventricle, the pulsation of the anterior horn of the lateral ventricle can be observed carefully to judge the range of anterior ventricular resection.
This study mainly de ned the range of SPTR of the frontal gliomas from an anatomical perspective.
Since the posterior boundary of the Hexahedron was de ned as the coronal suture and the lateral boundary as the inferior frontal sulcus, no motor or language disorders were produced in this study.
Because the frontal gliomas of this group were mainly limited to the frontal lobe, intraoperative techniques such as awake craniotomy, neuronavigation and image-guided surgery were not applied.

Conclusions
In this small series, the concept of the "Hexahedron" was shown to provide a safe, effective, and maximum extent of excision in the SPTR of the frontal gliomas.

Abbreviations
LGG   a skin incision; b the exposed area of the frontal lobe measures approximately 7x8cm2; c the olfactory nerve and the anterior skull base are exposed (white arrows); d the midline surface is exposed (white arrows); e the choroid plexus of the lateral ventricle is exposed (white arrows); f the resected area of the tumour and the frontal lobe measures approximately 7x8cm2.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. Table1.docx