The interhemispheric fissure—surgical outcome of interhemispheric approaches

Exposure of the anterior skull base is challenging due to strategic structures. The interhemispheric approach (IHA) has turned out to be a feasible technique. We report our experience with IHAs in patients with extraaxial lesions (EAL). We performed a retrospective chart review at a tertiary neurosurgical center between April 2009 and March 2020. We included patients with resection of EAL through IHAs concentrating on surgical technique, complete resection rate, postoperative outcome, and complications. Seventy-four patients resected by an IHA were included: 49 (66.2%) frontal (FIA), nine (12.1%) parietooccipital (PIA), and 16 (21.6%) frontobasal IHAs (FBIAs). Median age at time of surgery was 59 years (range 16–88 years), 47 (63.5%) female and 27 (36.5%) male. Complete resection rate was 83.8% (FIA 89.8%, PIA 55.6%, FBIA 81.3%). Rate of new minor deficits was 17.6%, rate of major deficits 5.4%, total rate 23.0%. 51 (68.9%) WHO°I meningiomas, ten (13.5%) WHO°II meningiomas, two (2.7%) WHO°III meningiomas, nine (12.2%) metastases, one (1.4%) sarcoma, and one (1.4%) local adenocarcinoma were resected. Total complication rate was 27.0%. Rate of major complications requiring intervention was 9.6%. Mean follow-up was 34.2 (± 33.2) months. In patients with lesions of the interhemispheric fissure, overall morbidity and complications are comparatively high. Extensions of IHAs with potential even higher morbidity are not necessary though; we support the use of standardized IHAs. Our findings suggest regular usage of relatively feasible IHAs for a satisfying outcome. Invasive, complicated, or contralateral trajectories were not needed.

As an alternative, but more as an extension, the frontobasal interhemispheric approach (FBIA) enables the view on the anterior skull base from the crista galli to the tuberculum sellae anteroposteriorly and from the midline to the sphenoid wing bilaterally. The approach has been described as safe, especially in terms of visual and pituitary stalk function [19,27,64]. A more subfrontal exposure is allowed.
More posteriorly, the parietooccipital interhemispheric approach (PIA) enables the resection of pathologies in the peritrigonal or periatrial region. This is challenging due to the depth of the region and due to strategically important structures [12,37,65].
The aim of this manuscript is to share our experience with a large series of different IHAs in patients with extra-axial oncologic pathologies at a tertiary neurosurgical university center. With the rise of new but complicated, technically challenging, poorly tested, and therefore potentially harmful approaches [9,[12][13][14]73], this study aims to highlight the sufficiency of standardized IHAs and thus the reduction of perioperative morbidity.

Study population and clinical parameters
We performed a noninterventional retrospective single-center study. Between April 2009 and March 2020, we screened the clinical documentation files and neuropathological records of patients who underwent surgery through an IHA for extraaxial tumors.
We analyzed clinical patient files for neurological symptoms, Karnofsky Performance Status Scale (KPSS), postoperative new neurological deficits, postoperative complications, reinterventions, and adverse events according to the Clavien Dindo scale (CDG). Radiological outcome parameters consisted of anatomic location as well as the extent of resection according to postoperative cranial magnetic resonance imaging (MRI) were recorded. We defined the total rate of postoperative new deficits/complications/reinterventions as the number of events divided by the number of patients. Postoperative deficits were defined as minor-to-moderate if they did not cause greater disability, whereas deficits associated with relevant loss of neurological function were defined as major. Deficits were defined as permanent if presented more than 3 months. Postoperative complications were classified in minor or moderate when no urgent intervention was needed and major in cases with subsequent intervention. A return to surgery was defined as an intervention.
We performed statistical analysis using STATA Version 13.1 (2011, StataCorp, College Station, TX). Data are shown as median and interquartile range or as mean and standard deviation; statistical significance was defined as p < 0.05.

Ethics approval
The local ethics committee of the Technical University Munich, School of Medicine, approved our study (231/20 S-EB). We conducted it in accordance with the ethical standards of the 1964 Declaration of Helsinki and its later amendments [72].

Study population
We included 74 patients who underwent surgical resection between April 2009 and March 2020. Median age of our patients at time of surgery was 59 years (range 16-88 years), with 47 (63.5%) female and 27 (36.5%) male patients. Seventeen patients (23.0%) had no preoperative symptoms; lesions were discovered incidentally during routine checkups or diagnostic workup for sinusitis or tinnitus. Fifty-seven (77.0%) patients were symptomatic, including cephalgia, nausea, diplopia, cranial nerve deficits, ataxia/imbalance, and further symptoms. The median preoperative KPSS was 90% (range 50-90), and the median postoperative KPSS 90% (range 0-100). Table 1 provides further detailed demographic  and clinical information. Tumor entities, location, and approach-related findings Histopathological analysis revealed 51 (68.9%) cases with WHO°I meningiomas, ten (13.5%) WHO II meningiomas, two (2.7%) WHO°III meningiomas, nine (12.2%) metastases, one (1.4%) sarcoma, and one (1.4%) local adenocarcinoma. Table 2 shows all entities and performed approaches in detail. FIA was mostly performed for falcine/parafalcine lesions (69.4%) and for lesions at the olfactory groove (22.4%). PIA was also used most commonly for falcine/parafalcine lesions of the middle or last third of the superior sagittal sinus (55.5%) but also for metastases in the cingular gyrus. The FBIA was additionally used for technically demanding basal midline meningiomas of the anterior skull base as well as for frontobasal falcine tumors. Table 3 displays the tumor location in relation of the described approaches.
Postoperative rate of new permanent minor-to-moderate neurological deficits was 17.6%. Most frequently, hemiparesis was observed in nine patients (12.2%), followed by psychomotoric disorders in two patients (2.7%). New postoperative hemianopsia was observed in one case. Regarding major deficits, postoperative hemiplegia was observed in two (2.7%), followed by dysarthria and neglect in one (1.4%) patient.
One patient with falcine meningioma of the anterior third of superior sagittal sinus committed suicide postoperatively. In another case with falcine meningioma, resected through a FIA, the ipsilateral pericallosal artery was injured and subsequently sacrificed, leading to a new postoperative permanent hemiparesis.
The overall postoperative complication rate was 27.0% (FIA 26.5%, PIA 22.2%, FBIA 31.3%) The rate of postoperative major complications was 9.6% The majority of   Proportionally, postoperative venous congestion occurred in 11.1% of the patients after a PIA, followed by 6.3% after a FIA (p > 0.05). None of them needed operative reintervention in the postoperative course. Venous congestion, shuntdependent hydrocephalus, and wound healing disorders were classified as moderate or minor complications (8.1%) as no immediate return to surgery was needed. (Fig. 1).

Discussion
Our series showed that the IHA is a technically and comparably safe operative technique. Despite new, but complicated and technically challenging approaches [9,[12][13][14]73], we could achieve complete resection (Simpson grades I + II and GTR) in 83.8% of our cases (exemplary cases 1-2) with standardized IHAs.
For decades, several authors have described alternative approaches to reach, e.g., suprasellar meningiomas or midline lesions of the anterior skull base as an alternative to the IHA. Median approaches, as presented in current series, might facilitate intraoperative orientation as anatomy is visualized in a straight line. The classic bifrontal approach allows good devascularization, especially in case of frontobasal midline meningiomas [1,2,8,50,52,56,61,62]. Still visualization of the anterior communicating complex and optic nerves might be laborious as they are hidden behind the tumor masses which might render this approach less suitable for tuberculum sellae meningiomas [11,36,40,52]. The FIA/FBIA might optimize the control of the anterior vascular structures and optic nerves but has been criticized for possible parenchymal retraction [39,46]. An adoption from the class bifrontal approach is the transbasal/subfrontal approach, a frontal lobe preserving approach initially described by Cushing [17,44,70] which enables superior devascularization with less brain retraction. Wide exposure and unilateral sparing of the contralateral frontal lobe have been described as some of the advantages [20,24,51,52,56,66], but pitfalls of that approach like opening of frontal sinuses and the invasivity with inferior border osteotomies sometimes not being sufficient enough to avoid brain elevation and making neurovascular dissection necessary have let more authors emphasize the advantages of dissecting the interhemispheric fissure for wider exposure. In addition, the risk for CSF leakage might be increased [15,26,42,52,[54][55][56]60]. In contrast, other authors advocate more technically demanding transbasal approaches to reach the anterior skull base [20,24,66]. Therefore, the FBIA enables a broader anatomical corridor of the interhemispheric fissure that might reduce frontal lobe retraction [19,20,27,56,63,69]. The FBIA might offer a better visualization if the boney skull base is infiltrated [19,20,27,56,63,69]. Nakamura et al. also showed a higher total resection rate (Simpson grade 1 or 2) through a median approach in their series of 82 OGMs: 91.2% (frontolateral) vs. 93.5% (bifrontal) [52]; however, they still advocate the less invasive frontolateral approach due to experienced lower morbidity rate and no mortality.
Lateral approaches allow for satisfactory and good visualization of the anterior circulation on both sides as well as superior access to the optic nerve, chiasm, and pituitary complex [1,11,35,49,52,74]. The classic pterional and frontolateral approach [35,61,74] offers wide visualization of parenchyma and skull base especially for lesions extending more laterally. As for intradural preparation, meticulous drilling and irrigation is necessary in order to avoid heat related damage to the optic structures [51,52,54]. Various modifications of these "working horses" in skull base surgery have been published and trended over recent years [4,16,32,33,57]. Purely endoscopic or endoscopy assisted resections via transnasal or transcranial routes have been popularized for some time offering minimal invasive visualization of the skull base region [49,59]. This trend has not peaked into a relevant paradigm shift as extent of resection, technical feasibility, and CSF leakage rates have been described to be unfavorable compared with classic skull base approaches [7,41,49,59]. Therefore, we do not see any reasonable indication to perform such techniques. Otherwise, if endoscopic approaches had led to any noteworthy advantages, such techniques would have been performed more regularly and spaciously.
However, regarding FIA, we observed postoperative hemorrhages and hematomas as noteworthy complications (12.0%). Three epidural, one subdural hematoma, and two intracerebral hemorrhages occurred. Brain retraction as a potential factor for postoperative intracerebral hemorrhages should be consequently reduced as much as possible. Furthermore, the craniotomy and the chosen trajectory should be carefully adapted to the planned approach in order to reduce retraction. Intraoperative meticulous hemostasis as well as closure and usage of tenting sutures, for example, may be taken into consideration as well, especially for the prevention of sub-and epidural hematomas [71,76]. The FIA remains a suitable approach for large meningiomas of the anterior cranial base, even with involvement of the anterior cerebral arteries into the posterior aspect of the tumor, due to an optimized view from superior-posterior and superior-anterior.
Regarding the PIA, one should notice that crucial vascular structures such as important central bridging veins, the vein of Trolard, and the Rolandic vein are located nearby. Preservation is mandatory during resection, even more than in FIAs, as more of the central region is exposed when more posterity is chosen. Those veins still limit the dural opening of the already narrow working space [43,58,68]. We also experienced such difficulties. This complication may be one of the main challenges of the PIA. The importance of preserving bridging veins is not negligible. The study of preoperative MRI to detect prominent bridging or anastomotic veins is essential for sufficient planning.
Major deficits, such as neglect, aphasia, and hemiplegia, respectively, could be considered as a result of exposure of the central region or the parietal lobe, too much brain retraction but also due to damage of frontal and central bridging veins, or ligation of the superior sagittal sinus. Aryan et al. showed that incidence of transient postoperative hemiparesis in infants appeared to be higher in those who required ligation of one or two parasagittal veins (44.6% versus 18.5%) in their series of 65 IHA in infants [5] (exemplary case 3). Spicer et al. described required coagulation and lysis of bridging cortical veins from the convexity to the sagittal sinus in 41.5% in their series of 53 cases of IHAs for mass lesions in childhood, which resulted in postoperative hemiparesis in 30% [67]. Exposure of the superior sagittal sinus is correlated with a better angle of view [3], but also increases the risk of injury of the sinus and the necessity of sacrificing parasagittal/ bridging veins, as seen above, so that especially in the middle and posterior segment the preservation of integrity of the superior sagittal sinus and related bridging veins is of utter importance for the postoperative outcome. In cases of postoperative venous congestion, a low-dose heparinization and further change to low-molecular-weight heparin should be taken into consideration. Especially in case of asymptomatic patients, we experienced no need of extensive postoperative heparinization. Recent retro-and prospective studies also showed higher radiographic incidence of postoperative cerebral venous sinus thrombosis than reported in retrospective studies [10,22,29,30]. They also advocated conservative treatment in absence of symptoms; collateral venous drainage is discussed as one of the main mechanisms for well thrombosis tolerance.
To conclude, exposure of superior sagittal sinus, the central region, sacrificing bridging veins, especially parietally, and retraction in case of suboptimal approach and trajectory planning has to be minimized to fully enable the advantage of this comparably cortex sparing technique.
As an extension of the conventional IHA, FBIA can be chosen if the right indication is given and if pathoanatomical conditions require this more extensive approach, such as for large midline meningiomas or craniopharyngiomas [19,20,27,56]. Primary described by Suzuki et al. as a variation of the bifrontal anterior interhemispheric approach for 3rd ventricle tumors and anterior communicating artery aneurysms, the FBIA has been extended and modified in the last decades to expose the anterior skull base from a more subfrontal view [63,69]. The FBIA has all advantages of the classic anterior subfrontal approach and of the FIA by using the anatomical corridor of the interhemispheric fissure. Frontal lobe retraction can be minimized if the fissure is sufficiently divided. This kind of approach can expose the complete tumor height. Ganna et al. could show satisfactory postoperative outcomes in their series of resected tuberculum sellae meningiomas by FBIA, especially no new visual deficits were observed in their patient cohort, which reflects the outcome of our patient group [27].
Surgically, early identification of the optic apparatus is essential; in most cases, the optic nerves are found displaced posteriorly laterally. After initial tumor debulking further coagulation may be done cautiously as carotid artery often could be placed medial to the displaced optic nerves. Due to already mentioned advantages of exposure and point of view by the FBIA, retraction of the optic nerves is normally not necessary. However, attention must be paid to any arachnoid plane as it serves as a natural layer between the tumor and the neurovascular structures; mobilization and complete tumor removal are possible but may be refrained in case of greater adherence in order to maintain visual function. In some cases, the chiasmatic and lamina terminalis cistern has to be exposed more extensively. Afterwards, the anterior communicating complex will come into view and enables more intraoperative control. Unroofing the optic canal and incision of the falciform ligament may be required in case of tumor infiltration into the optic canal and in case of preoperative visual impairment. Such decompression is not mandatory in every case as drilling the optic roof, and the tuberculum sellae may cause additional CSF leak.
To sum it up, we emphasize detailed exposure and wide interhemispheric dissection prior to sometimes too early tumor resection in order to expose, view, and therefore control all necessary structures.
In our series, FBIA is associated with highest approachrelated complication rate ( Regarding complications, cerebrospinal fluid (CSF) leaks are considered the most common ones through such frontobasal techniques, similar to transbasal approaches, especially after dural reconstruction [15,26]. However, we experienced no postoperative CSF leak at all, satisfactory results compared with the findings of for example Obeid [42,56,60], but one should notice that they referred to more transbasal approaches than we did. The basic surgical technique of the FBIA is already described in detail by other authors before [27,64]. In fact, we did not use any noteworthy alterations of known techniques, but we advocate regular usage of a navigation system, if applicable, as frontal sinus can be detected preciously and possibly be spared out during craniotomy. We experienced that in some cases, full bony basal exposure is not that mandatory as interhemispheric dissection and the angle of view from anterior to superior or, after new table and microscope positioning, even posteriorsuperior, is satisfactory. Of course, in case of greater lesions with even infiltration of the anterior skull base, full exposure is necessary. In such cases, besides cranialization of the sinus, meticulous dural reconstruction and skull base coverage must be performed, whereas we prioritize priory preparation und usage of a galea-periosteum-flap. We moved away from using fascia lata in the first place as we experienced more CSF leaks than with the flap, which may be based on insufficient vascularization. Necrosis and thus also source of superinfections may be further consequences. However, this just reflects our own institutional experience, whereas techniques of anterior skull base reconstruction have been extensive discussed and algorithms proposed [28,31,34].
Anyway, comparison of complication rates is difficult as other authors referred to specific entities, outcome parameters or were small sample sized [20,27,45,64].
Compared with the other approaches, we observed a noteworthy frequency of infection-related complications by FBIA. Incidence of infection is relatively low [15,26,28,45,60,62]. They may occur in the case of larger tumors and longer operative times requiring surgical revision with bone flap explantation and plastic flap reconstruction, as with one of our patients (6.3%). Accidental opening of frontal sinus and consecutive contamination, especially in the case of sinusitis, must be taken into consideration (Figs. 1, 2, 3 and 4). Careful cranialization and sinus obliteration with removal of the mucosa, if needed, should be performed to minimize the postoperative infection rate. In the case of prominent sinus, sinusitis, or other active nasosinal infections, another approach may be considered.

Study limitations
As it is a retrospective case series, it is not possible to draw causalities with respect to clinical outcome. Nevertheless, we implemented detailed clinical examination, including scores on functional performance, and a standardized follow-up protocol based on a certified neurooncological board into our clinical workflow. Nevertheless, the current study has some noteworthy limitations. Besides of its retrospective nature, the analyzed patient collective suffers from certain aspects of heterogeneity. First, the PIA is limited to 9 on a total of 74 which could lead to variability in the results. We decided to include the PIA into analysis as all aspects of interhemispheric approaches should be reflected and as basic surgical techniques do not differ exceptionally between FIA and PIA.
Secondly, variability of pathologies included could lead to further heterogeneity as majority of cases were meningiomas. With aiming at the approach related complications, we decided to focus on extrinsic lesions, as manipulation techniques seemed to be more similar, so complications could be reduced on the approaches as good as possible.

Conclusions
In patients with lesions of the interhemispheric fissure, overall morbidity and complications are comparatively high. Modifications or extensions of IHAS with a potential even higher morbidity are not necessary; we support the use of standardized IHAs for a variety of entities. Our findings suggest regular usage of relatively feasible IHAs for a satisfying outcome and maximal extent of resection. Invasive, complicated, or contralateral trajectories were not needed.
Availability of data and material Data and material are not publicly available.

Compliance with ethical standards
Conflict of interest Jens Gempt (JG) and Bernhard Meyer (BM) work as consultants for Brainlab (Brainlab AG, Feldkirchen).
In addition, BM works as a consultant for Medtronic, Spineart, Icotec, Relievant, and Depuy/Synthes. In these firms, BM acts as a member of the advisory board. Furthermore, BM reports a financial relationship with Medtronic, Ulrich Medical, Brainlab, Spineart, Icotec, Relievant, and Depuy/Synthes. He received personal fees and research grants for clinical studies from Medtronic, Ulrich Medical, Brainlab, Icotec, and Relievant. All this happened independently of the submitted work. BM holds the royalties/patent for Spineart. All named potential conflicts of interest are unrelated to this study. There are no further conflicts of interest regarding the other authors.
Ethics approval The local ethics committee of the Technical University Munich, School of Medicine, approved our study (231/20 S-EB). We conducted it in accordance with the ethical standards of the 1964 Declaration of Helsinki and its later amendments [72].
Informed consent The requirement for written informed consent was waived by the ethics committee.
Code availability Not applicable.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.