Surgical excision of large-to-giant petroclival meningiomas focusing on the middle fossa approaches: The lessons learnt
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.241354
Source of Support: None, Conflict of Interest: None
Introduction: Petroclival meningiomas are based on or arising from the petro-clival junction in upper two-thirds of clivus, medial to the fifth cranial nerve. This study focuses on the surgical experience in resecting large-giant tumors >3.5 in size predominantly utilizing middle fossa approaches.
Keywords: Anterior petrosectomy, giant, outcome, petroclival meningiomas, subtemporal approach, surgery, trans-sylvian approach
Petroclival meningiomas, due to their location, multi-compartmental extent and intricate relationship with critical neuro-vascular structures were considered as inoperable lesions not long ago. Only one case of total successful removal of a petroclival meningioma had been reported before 1970, and the operative mortality in those times usually exceeded 50%. With the advent of microsurgical techniques, different skull base approaches, stereotactic radiosurgery and advancements in anaesthesia and post-operative care, there has been a tremendous improvement in the outcome of these patients.
Al Mefty et al., defined true petroclival meningiomas as tumors based on or arising from the petro-clival junction in the upper two-thirds of the clivus, medial to the fifth cranial nerve. Large petroclival meningiomas are defined as having a size of more than 3 cm., Sphenopetroclival meningiomas are the largest of these lesions and usually have some extension into the middle cranial fossa, namely into the Meckel's cave, cavernous sinus, suprasellar region, temporal base or the sphenoid wing, while a lateral extension beyond the internal auditory meatus (IAM) is variable.
Petro-clival meningiomas progressively increase in size and ultimately cause brain stem compression and death. Van Havenbergh et al., have published the largest series on the natural history of petroclival meningiomas. They found that 76% of untreated petroclival meningiomas showed radiographic growth over a mean period of 82 months. 63% of these patients whose tumors grew, had a significant functional decline proving that, except in the circumstances when a very small tumor is present and/or when the patient's general health is poor, these tumors should be resected.
Gross total excision of the petroclival meningiomas offers the only definite chance of cure. However, in today's era, the trend has shifted from radical excision of the tumor to near-total and sub-total excision of the tumor, especially in large (>3 cm) and giant tumors (>4 cm) to reduce the post-operative morbidity and preserve the quality of life of the patients. Surgery may be followed by stereotactic radiosurgery (SRS) if there is radiological evidence of progression of the disease. It has been observed that SRS is not required in all patients with sub-totally resected tumors, and in most patients, observation and follow up is sufficient.
There have been a number of different skull base approaches described in the literature for the management of petroclival meningiomas but controversy regarding the ideal approach exists till date. In this article, the feasibility of surgical resection of large and giant petroclival meningiomas, predominantly utilizing the middle fossa approaches, including the half-and- half approach as well the Kawase's anterior petrosectomy, has been emphasized, and the outcome compared with their excision using the retrosigmoid suboccipital approach. The clinico-radiological and surgical nuances in the management of extensive petroclival meningiomas are also discussed.
We included 33 patients having a large or a giant petroclival meningioma (those with a size >3.5 cm), operated at our tertiary care institute between 2009 and 2018 in this study. The clinical features, preoperative radiological details, operative findings, the postoperative clinical course and clinico-radiological details at the follow-up visit were reviewed for each patient. At admission and on the day before surgery, the clinical symptoms and the presence of cranial nerve deficits, long tract signs, and cerebellar signs were documented. The House-Brackmann grading of facial nerve weakness was used to grade the preoperative and postoperative facial nerve function.
Preoperatively, all patients underwent a contrast enhanced computed tomography scan (CT) and a magnetic resonance imaging (MRI) for a detailed evaluation of the size, extension and contrast enhancement that was characteristic of the tumor. The tumor extension into the adjacent middle and posterior skull base regions like the Meckel's cave, cavernous sinus, sphenoid sinus, internal auditory meatus (IAM), and along the tentorium and the clival region were noted.
The relationship of the tumour with important neurovascular structures were noted as this aided in the safe removal of the tumor. MR angiography and digital subtraction angiography (DSA) were performed in patients in whom encasement of major arteries was suspected on the initial imaging to assess the tumor vascularity, the course of the major vessels encased within the tumor, and to understand the related vascular cross flow for surgical planning. Preoperative tumor embolisation was not done in any of the patients.
We subclassified petroclival meningiomas based on their size into two groups; Group A with their size between 3.5 cm and 5 cm, and Group B with their size above 5cm. The surgical outcomes in both the groups were compared. The surgical approaches were planned depending upon the location and extension of the tumors. In this study, a significant number of the patients were operated using the middle fossa approach including the anterior petrosectomy, and the half-and-half approach. Those tumors, predominantly located below and posterior to the internal auditory meatus (IAM) in the cerebellopontine (CP) angle and the inferior cerebellomedullary cistern were operated by the retromastoid approach or a combination of the retro-mastoid and middle fossa approaches.
In the postoperative period, all patients underwent a contrast enhanced CT scan for the documentation of the presence of any residual tumor on the day of surgery. On the basis of postoperative images, the extent of resection was described as ‘gross total' (no residual tumor), ‘near total' (<10% residual tumor remaining) and ‘subtotal' (>10% residual tumor remaining) resection. The postoperative complications, that is, cranial nerve deficits, limb weakness, sensory deficits, cerebrospinal fluid (CSF) leak and seizures were noted.
Glasgow outcome scale (GOS) was used to quantitatively score the postoperative neurological outcome. On follow up visit, the clinical status of the patients and the differences in the neurological status, as compared with the preoperative status, were noted. The radiological (contrast enhanced CT/MRI) imaging was done at 6 weeks and at regular intervals to document the recurrence or progression of meningiomas. The patients were followed up at 6 weekly intervals for three months; then at 3 monthly intervals for one year, and then, at six monthly intervals thereafter, on an outpatient basis. Adequate clinical and radiological follow up was available in 27 patients and 6 patients were lost to follow up. The mean follow up duration of our study group was 35.77 months (range, 1-106 months).
Headache was the most frequent symptom, present in 22 out of 33 (66.7%) patients. 20 out of 33 (60.6%) patients presented with the clinical features of a cerebello-pontine angle syndrome. Fifth cranial nerve was the most commonly involved cranial nerve in 19 out of 33 (57.6%) patients, followed closely by the eighth cranial nerve in 18 patients (54.5%). The extra-ocular motility (related to cranial nerves III, IV and VI) was impaired in eight patients. Two patients presented with seizures. One of our patients presented with pathological laughter that improved following excision of the tumor [Table 1].
Characteristics of the tumor
Only large (>3.5 cm) and giant (>4 cm) petroclival tumors were included in our study. 17 (51.5%) patients had a tumor ranging from 3.5 cm to 5 cm in its greatest dimension, while the rest had a tumor larger than 5 cm in its greatest dimension on radiological examination. 25 (75.8%) patients had their tumour extending into both the supratentorial and infratentorial compartments. The tumour extended into the Meckel's cave in 25 (75.8%), the cavernous sinus in 17 (48.4%), the sphenoid sinus in 12 (38.7%) and the suprasellar area in 12 (38.7%) patients. 31 (93.9%) patients had their tumor crossing the midline in the premedullary, prepontine and interpeduncular cisterns posterior to the clivus. In 20 (60.6%) patients, the tumour was extending below and posterior to the IAM; while in 13 patients (39.4%), the tumor was located above and anterior to the IAM. 11 (33.3%) patients had brain stem hyperintense signal on T2 weighted MRI suggestive of brain stem edema. 15 (45.5%) patients had encasement of either of the three vessels, namely the basilar artery (BA), posterior cerebral artery (PCA) or internal carotid artery (ICA). 22 (66.7%) patients had hydrocephalus, out of which 9 had to undergo a cerebrospinal fluid (CSF) diversion procedure in the form of a ventriculo-peritoneal shunt in the pre-operative period due to the extremely poor neurological status of the patients at admission [Table 2].
Surgical approaches used
Based upon the location of the major bulk of the tumor in relation to the internal auditory meatus, four different approaches were utilized for the removal of petroclival meningiomas. A temporal craniotomy with a zygomatic osteotomy and an anterior petrosectomy (Kawase's approach) was the most commonly used approach, that was utilized in 16 (48.48%) patients. Other approaches included the half-and-half (trans-Sylvian with subtemporal) [n = 6, 18.18%]; frontotemporal craniotomy with orbitozygomatic osteotomy [n = 1, 3%] and retromastoid suboccipital craniectomy (RMSO) [n = 7, 21.21%], respectively. In 2 (6.06%) patients, a staged anterior petrosectomy with RMSO; and, in 1, a staged presigmoid with half-and-half approach was used. In 2 (6.06%) patients, the staged procedure of anterior petrosectomy followed by the RMSO approach was used. In 1 of these 2 patients, the anterior petrosectomy approach had to be abandoned due to excessive bleeding from the region of the cavernous sinus and the tumor was removed by the RMSO approach on the next day; while in the other patient, the tumor was giant, having both middle fossa and posterior fossa components, and had to be removed using both the approaches in a staged manner. In one patient, a combined half-and-half approach as well as a RMSO craniectomy was used. The anterior petrosectomy approach had the maximal rate of total tumor resection with gross total resection being achieved in 8 patients and near-total resection in 7 patients. One patient initially operated in 1998 developed a recurrence in 2010 and was reoperated by the anterior petrosectomy. Another patient initially operated in 2009 by the presigmoid approach was re-operated in 2011 by the middle fossa half-and half approach for the residual tumor in the sellar-suprasellar region extending to the cavernous sinus and temporal base [Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7], [Figure 8], [Figure 9], [Figure 10].
Extent of resection
Gross total excision was performed in 12 (36.36%) patients, followed by near-total excision in 15 (45.45%) patients and a subtotal excision in 6 (18.18%) patients [Table 3]. The reasons for performing only a subtotal excision in 5 patients were: a very large size of the tumor with extension from the sphenoid ridge to the internal auditory meatus, along with the involvement of the sphenoid sinus and/or the cavernous sinus; the firm consistency of the tumor; and, its dense adherence to the brain stem, jugular foramen and/or the cranial nerves.
In one patient, >20% of the tumor was left behind during the performance of the middle fossa half-and-half approach, since most of the tumor bulk in this patient was below and posterior to the IAM. A second stage RMSO to ensure maximal tumour removal could not be done as the patient refused a second stage surgery.
Post-operative morbidity and mortality
6 and 14 patients had the post-operative GOS scores of 4 and 5, implying that 20 (60.6%) patients had a good functional outcome postoperatively, in a mean follow-up duration of 35.77 months. There were six deaths in our study, of which 3 patients died in the perioperative period and the other three succumbed in the follow up period. Of the three perioperative deaths, two patients were in an altered sensorium at admission; one of them succumbed to cerebrospinal fluid leak from the main wound and the development of meningitis; and, the other developed ventriculitis after a repeated ventriculo-peritoneal shunt failure for hydrocephalus. The third patient died due to brain stem perforator injury in the immediate postoperative period.
In the three patients with delayed mortality, one patient died 10 months after surgery as a consequence of progressive brain stem compression due to the growth of residual tumor despite the patient having received adjuvant radiotherapy; another patient who had lower cranial nerve paralysis with associated poor gag and swallowing reflexes succumbed at 5 months after surgery due to aspiration pneumonitis, following aspiration of oral secretions during a seizure episode while she was being nursed under domiciliary care; and, the third patient, discharged on a tracheostomy due to his persistent lower cranial nerve paresis and poor clearance of respiratory secretions, died 7 days after discharge owing to aspiration pneumonitis.
7 (21.2%) patients developed a transient impairment of extra-ocular motility in the immediate postoperative period. Four of them developed complete ophthalmoplegia, one developed third nerve paresis, one developed sixth nerve paresis and the remaining one developed both fourth and sixth nerve paresis. 6 (18.18%) patients developed permanent external ocular motor paresis. 2 of them developed complete ophthalmoplegia, 2 had fourth nerve paralysis and the remaining 2 had third and sixth nerve paresis, respectively. 2 of these patients had only a partial improvement in the third nerve function, in the postoperative period.
5 patients had improvement in their fifth nerve symptoms after surgery. One patient developed worsening in the motor fifth nerve function in the form of an increase in temporalis muscle atrophy and dysfunction, one patient developed new onset transient facial hypoesthesia (fifth nerve involvement), while another had persistent facial hypoesthesia postoperatively.
7 (21.2%) patients developed transient facial nerve paresis in the post-operative period, of which one had contralateral upper motor neuron (UMN) type paresis with hemiparesis and the rest had ipsilateral lower motor neuron (LMN) type of paresis. One patient developed permanent facial nerve paralysis (grade 4). 2 patients had worsening of facial nerve functional status from the House Brackmann grade 2 to grade 4 in the postoperative period while 4 patients had improvement in their facial nerve function. None of the patients complained of post-operative new onset hearing loss, though their objective testing [using pure tone audiometry (PTA)] was not done in the post-operative period. 7 (21.2%) patients developed a transient lower cranial nerve paresis in the postoperative period, while 1 patient developed a persistent (continuing for a prolonged period as detected on the follow-up visits) lower cranial nerve palsy. 4 patients had a postoperative improvement in their lower cranial nerve function.
10 (12.1%) patients developed transient contralateral hemiparesis postoperatively, which gradually improved completely and none of the patients had a persistent hemiparesis in the follow-up assessment. 7 (21.2%) patients developed CSF leak from the wound which was managed conservatively. One patient developed meningitis in the postoperative period that recovered uneventfully with intravenous antibiotics. One patient developed pneumonia, which resolved with intravenous antibiotics and chest physiotherapy. 2 patients were readmitted in the follow up period with multiple episodes of seizures, which were managed by stepping up of the anti-epileptic drugs.
6 patients had to be tracheotomised after surgery, of which 3 required a tracheostomy due to their poor preoperative neurological status, and 3 due to lower cranial nerve involvement and inability to clear secretions. In three of these patients, the tracheostomy port could be successfully closed in the follow up period. [Table 3] summarizes the surgical outcome in terms of extent of resection and postoperative morbidity/mortality.
One patient underwent an endoscopic third ventriculostomy (ETV) in the postoperative period for hydrocephalus with a persistent pseudomeningocoele. One patient underwent a lumbo-peritoneal shunt in the postoperative period for CSF rhinorrhoea. The latter condition resolved following placement of the shunt.
A patient initially operated in 1998 with gross total excision, had a local recurrence with mass effect and was operated again in 2010. As gross total excision was performed again and the biopsy of the tumor showed a grade 1 meningioma, the patient has not received radiotherapy. In 2 patients, the tumor progressed in size after the initial surgery. One of these patients received postoperative radiotherapy, and the other patient was again operated after 2 years (he was initially operated in 2009 by the pre-sigmoid approach, and later on, in 2011, by the middle fossa approach). The biopsy report in all the 33 patients was consistent with a grade 1 meningioma.
Rationale for the functional classification for excision of petroclival meningiomas
The uniqueness of our study lies in this being a surgical experience with a relatively long-term follow up (mean: 35.77 months), where petroclival meningiomas of a large (>3 cm) or giant size (>4 cm) were resected. The middle fossa approaches rather than the posterior fossa approaches were predominantly applied for the successful excision of these lesions. ,,
There have been numerous attempts described in literature to classify petroclival meningiomas. All these classifications are helpful in deciding the optimal approach to excise the tumour but none of them is universally applicable to all varieties of petroclival meningiomas, irrespective of their size and extent.,,, In this study, a functional classification was utilized that helped in deciding the approach to be adopted. This classification divided these tumors into four categories based upon their extent of spread: 1) Tumors confined to the anterior petrous region medial to the IAM; 2) tumors that extended predominantly into the middle cranial fossa reaching upto the sphenoid wing anteriorly, and often involving the lateral wall or even the interdural and intra- cavernous parts of the cavernous sinus and the suprasellar region, with the posterior fossa extension remaining mainly superomedially to the IAM; 3) tumor mainly reaching across the midline anterior to the brain stem and also extending to the middle cranial fossa but remaining medial and superior to the IAM in the posterior fossa; and, 4) tumors being confined mainly to the posterior fossa, extending from the petrous apex to the cerebellopontine and cerebellomedullary cisterns, as well as the region lateral to or below the IAM, and occasionally extending into the jugular foramen. The two middle fossa approaches were used in the first three categories where there was a predominant middle fossa component of the tumor, with the posterior fossa component mainly confined to the region of the petrous apex and the part of the petrous bone that was predominantly superomedial to the internal auditory meatus. The frontotemporal craniotomy (with or without orbitozygomatic osteotomy) and half-and- half approach was used in the group 1 and 2 where the tumor was more anteriorly placed, with the middle and posterior cranial fossae components straddling and eroding the petrous apex and often extending to the suprasellar region. Group 3 tumors with a predominant posterior fossa component (along with middle fossa component) that were crossing the midline anterior to the brain stem, and which were mainly situated superomedial to the IAM were resected utilizing the Kawase's anterior petrosectomy. Group 4 tumors that were mainly confined to the posterior fossa and also extended laterally and below the IAM were resected utilizing the retromastoid suboccipital craniectomy. Occasionally, a combination of these approaches, either concurrently under the same anaesthesia or in a staged manner, were utilized.
The classical approach to resecting extensive/large petroclival meningiomas with a major extension into the middle fossa region is the anterior petrosectomy, also known as the extended middle fossa approach or the Kawase's approach. This approach was initially described by Bochenek and Kukwa, and later on, modified by Kawase.,,, This approach consists of approaching the petroclival region via the middle fossa and drilling the petrous apex through the Kawase's quadrilateral, to connect the middle and the posterior cranial fossae, so that the tumor, straddling across the petrous apex into the middle and the posterior cranial fossae, is adequately exposed and resected. We used this technique in 16 of 33 (48.48%) cases as the primary approach and in 2 (6%) others in combination with other approaches. The gross-total resection rate and the near-total resection rate using this approach in these extensive petroclival meningiomas was 66.67% and 46.67%, respectively. While utilizing this approach, the following technical points were taken into account for complication avoidance. The superficial temporal fascia is attached to the zygomatic arch while the temporalis muscle along with the deep fascia lies below the zygomatic arch and is attached onto the ramus of mandible. The zygomatic arch needs to be resected to ensure a good exposure of the middle fossa base. This is done by incising the superficial temporal fascia over the zygomatic bone, drilling the zygomatic arch at its anterior and posterior ends, and retracting it downwards with the temporalis muscle and its deep fascia. The temporomandibular joint is located just posterior to the inferior margin of the root of zygoma. Care should be taken not to expose it, as an inadvertent opening of the joint capsule may result in difficulty in jaw opening in the postoperative period. The root of the zygomatic arch is the landmark for the petrous apex. After a temporal craniotomy, the groove of the middle meningeal artery leads on to the foramen spinosum. Further extradural temporal lobe retraction is only possible following division of middle meningeal artery. The groove of the greater superficial petrosal nerve is the surface landmark for the bony petrous canal harbouring the internal carotid artery. The sharp dissection of the temporal lobe dura from the cavernous sinus dura, from the superior orbital fissure to the Gasserian ganglion More Details and the origin of the mandibular nerve, facilitates extradural temporal lobe retraction, exposing the extradural maxillary and mandibular nerves, the Gasserian ganglion, the lateral wall of the cavernous sinus and the Kawase's rhomboid (consisting of mandibular nerve and Gasserian ganglion anteriorly; the bony groove of the greater superficial petrosal nerve laterally, the edge of the superior margin of the petrous bone with the superior petrosal sinus medially, and the arcuate eminence harbouring the superior semicircular canal posteriorly). Drilling the Kawase's rhomboid exposes the temporal as well as posterior fossa dura and the lateral wall of the cavernous sinus. The opening of the temporal dura, slight retraction of the temporal lobe while preserving the vein of Labbe, dividing the tentorium posterior to the point of entry of the IVth nerve at the tentorial incisura, and then dividing the superior petrosal sinus, permits the conversion of the temporal base and the posterior fossa into a single cavity, thus adequately exposing the petroclival meningioma. The cavernous part of the tumor may be removed extradurally from the triangle between the ophthalmic and maxillary divisions, as well as the triangle between the maxillary and mandibular divisions of the trigeminal nerve. Maintaining an arachnoidal plane, the tumor is gently lifted from the brain stem preserving the neuraxis as well as the Vth nerve, posterior cerebral artery and the anterior choroidal artery traversing in the ambient cistern lateral to the brain stem. The tumor anterior to the brain stem and shifting the basilar artery may also be removed using this approach. The simultaneous neurological involvement of the Vth and VIIth nerves with this tumor may lead to neuroparalytic keratitis mandating a tarsorrhaphy.
In the half-and-half approach, a basal subtemporal and a wide trans-Sylvian dissection is simultaneously undertaken utilizing a frontoemporal craniotomy and zygomatic osteotomy. After the wide trans-Sylvian dissection, the temporal pole is retracted posteriorly exposing the suprasellar area, the tentorial incisura and the ambient cistern. The tentorium is divided transversely from the incisura, posterior to the point of entry of the IVth nerve, gaining access to the tumor from the middle to the posterior fossae as well as the cavernous sinus. When the latter approach is compared to the Kawase's anterior petrosectomy, a greater access to the suprasellar region and the sphenoidal wings is obtained using this approach; while a greater access to the posterior aspect of the ambient cistern, and the anterior aspect of the brain stem upto the basilar artery is obtained utilizing the Kawase's approach., Nair et al., have used this approach to effectively manage petroclival meningiomas and have reported excellent results.
The advantages of the two middle fossa approaches are that they provide the most direct trajectory and a good exposure of the tumors straddling across the petrous apex, especially those in the region of the temporal base, namely the posterior cavernous sinus and the sphenoid wing.,, They also provide a good exposure along the posterior surface of the clivus across the midline anterior to the brain stem. Since most of the extensive petroclival meningiomas in our series had crossed the midline anterior to the brain stem and were often straddling across the middle and posterior fossae, these approaches were the main workhorse approaches used in our series. By directly gaining a wide exposure to the surface of the tumor, its devascularisation and piece-meal decompression was possible, permitting the tumor capsule to roll over into the cavity created, thus preserving the arachnoidal plane between the tumor and the neuraxis in the vicinity. As cranial nerves are not handled during these approaches, except for the intra- and extradural parts of the Vth nerve, there is often preservation of function of the cranial nerves in the vicinity of the tumor.
The posterior fossa approach utilizing the retrosigmoid suboccipital craniectomy approaches the petrous apex by retracting the cerebellum downwards and medially, directly accessing the cerebellopontine angle cistern and the Vth cranial nerve that is traversing the subarachnoid space from the Gasserian ganglion at the petrous apex to its root entry zone at the brain stem. An adequate cerebrospinal fluid drainage from the cisternal magna and the inferior cerebellomedullary cistern helps in improving cerebellar retraction. However, the approach cannot be used to reach the part of the tumor that is attached to the clivus anterior to the brain stem. Goel et al., have described the use of this approach in their series of 28 cases of petroclival meningiomas but the predominant location of the tumors in their series was in the posterior fossa that had considerably shifted the brain stem towards the contralateral side, thus creating space for the trajectory of approach. Gupta et al., have used this approach to access the middle fossa part of the tumor by drilling the petrous apex intradurally, and accessing the supratentorial portion of this dumbbell tumor.
A pre-sigmoid approach was adopted for a single case of a petroclival meningioma that had extended into the cerebellopontine angle and the cerebellomedullary cistern. This approach had the advantage of devascularizing and decompression of tumor prior to the encountering of the cranial nerves. A C-shaped post auricular incision was made just 3 cm behind the post-auricular sulcus. A musculo-periosteal flap was reflected anteriorly along with the pinna until the posterior wall of the external auditory canal. Then, prior to the drilling of mastoid, the bony landmarks, namely the inferior temporal line, the spine of Henle and the mastoid tip were identified. The drilling was started at the MacEwan's triangle and the outermost cortical air cells were removed. Superiorly, the middle cranial fossa dura was delineated, and posteriorly, the sigmoid sinus was exposed. Then, further exenteration of mastoid air cells was done along with removal of retrolabyrinth air cells and retrofacial air cells to define our boundaries- middle cranial fossa dura superiorly, lateral semi-circular canal medially, the pre-sigmoid dura posteriorly, the sinodural angle posterosuperiorly, the bony canal harbouring the facial nerve anteroinferiorly, and the posterior semicircular canal posteromedially. The vertical segment of facial nerve anteroinferiorly was delineated with the intact Fallopian canal More Details till the anterior end of diagastric groove. The presigmoid dura was opened parallel to the sigmoid sinus, just short of the jugular bulb and was connected to the horizontal temporal dura after ligating and cutting the superior petrosal sinus. The cranial nerves fifth, seventh, eighth and lower cranial nerves were adequately exposed using the presigmoid approach. The advantage of this approach was that a major portion of the tumor was debulked prior to the encountering of the cranial nerves and other neurovascular structures at the cerebellopontine angle.
One disadvantage of this approach is that occasionally, the meatal part of the facial nerve may have to be mobilized to reach the petrous apex. A more extensive drilling, which involves drilling of the semicircular canal and the labyrinth causes loss of hearing. Injury to the sigmoid sinus may result in excessive haemorrhage, air embolism, significant brain edema or cortical venous thrombosis. This approach, however, achieves a good clearance of lesions extending from anterior to the IAM until the petrous apex.
For a complete as well as a safe resection of these extensive petroclival tumors, the caveats include the utilization of multiple corridors to the middle fossa and the brain stem region; the utilization of multiple approaches either under the same anesthesia or in a staged manner; whenever a plane of dissection between the tumor and the neuraxis is not available, leaving a thin rim of tumor over brain stem rather than attempting to excise it completely (as occasionally, a perforator compromise causes a significant neurological deterioration, as occurred in one of our patients); and, the occasional utilization of adjunctive therapy to prevent growth of the remnant tumor left in situ.
Role of stereotactic radiosurgery (SRS)/preoperative embolisation in petroclival meningiomas
Radiosurgery has a debatable role in the management of petroclival meningiomas. The best chance of cure in petroclival meningiomas is gross total resection, which is often feasible in smaller sized tumors but carries the risk of cranial nerve morbidity/mortality in large petroclival meningiomas. Many studies have shown that subtotally resected tumors do not have a significant risk of recurrence even when post-operative radiation has not been administered. The reasons for this observation are not clear though decrease in the vascularity of the tumor and its proliferative cell population have been proposed as the mechanisms responsible for further lack of growth of the tumor. Though most of the studies on the role of SRS in petroclival meningiomas, both as a primary and adjuvant treatment, show that it decreases the risk of progression in size and recurrence of the tumor, respectively, especially in small and medium sized tumors, there are many important caveats to this statement., Most of these studies have no long-term follow-up; sometimes, the skull base meningiomas exhibit an aggressive behaviour following radiosurgery; surgery for the tumor after it has undergone radiation therapy becomes difficult due to the extensive peritumoral fibrosis and the change in consistency of the tumor to a more solid one; and, primary treatment with radiosurgery may not be suitable for large petroclival meningiomas, especially in close relation to sensitive neurovascular structures, as seen in the present study., Furthermore, the risk of radiation induced malignancy has also been reported.,,,,,,
Comparison of results reported in literature
Gross total resection (GTR) was performed in 12 of 33 (36.36%) patients in our study, which was comparable to other major series. Thus, Al Mefty et al., reported a GTR in 64.6%, Seifert et al., in 37%, Nanda et al., in 28%, and Natarajan et al., in 32% of their patients. However, most of the other series have not specified the size of their tumor while we only included those patients in our study who had a tumor size greater than 3.5 cm. In our series, there were a total of six (18%) deaths, of which 3 (9%) occurred in the peri-operative period and 3(9%) occurred in the follow up period. Majority of them occurred in poor grade patients with poor respiratory reserve and lower cranial nerve paresis. Goel et al., reviewed 28 patients of petroclival meningioma operated by the RMSO approach and reported a mortality in 7.1% patients. New onset cranial nerve morbidity in our series was 42.42% which was comparable to that reported in the other series (38% by Al Mefty et al., 37.5% by Feng Xu et al., 44% by Nanda et al., 31% by Seifert et al., 30% by Bambakidis et al., 32.14% by Goel et al., and 40.14% by Little et al.) [Table 4].
A rare presentation of petro-clival meningiomas
One of the patients in our series presented with sudden outbursts of spontaneous, uncontrollable and inappropriate laughter along with hearing loss, numbness over the face, and cerebellar symptoms. Pathological laughter as a presenting feature may have localizing value, as has been reported in the literature., As reported in other case reports, it immediately resolved after the surgery in our patient also. The pathophysiology of this manifestation remains incompletely understood though compression over the brain stem and the medial temporal lobe structures has been proposed as one of the mechanisms. Since pathological laughter does not occur in every case of petroclival meningioma, other still unknown factors may play a role in its genesis.
Management of postoperative neurological deficits
Optimum management of persistent cranial nerve palsy is essential to improve the quality of life of the patients in the post-operative period. In the presence of diplopia due to extraocular paresis, an intermittent placement of an eye patch helps, as it changes the vison to a uniocular one and eliminates diplopia in the interim period before it recovers completely. Persisting ptosis due to IIIrd nerve palsy may be resolved by an eyelid-lift surgery. Management of seventh nerve paresis in the post-operative period consists of protecting the eye from corneal desiccation and ulceration, in addition to improving the facial asymmetry. Other than tarsorrhaphy, which is considered a temporary and a less cosmetic procedure, better options include gold implants, which may aid in eye closure, and facio-hypoglossal nerve anastomosis, which has been shown to give excellent results even after two years of iatrogenic injury., Lower cranial nerve paresis is usually temporary in the post-operative period, which resolves after a few months. It is managed with nasogastric feeding or by institution of a feeding jejunostomy. Occasionally, the patient may require a tracheostomy. In some recalcitrant cases, where it is showing no signs of improvement and there is persistent and recurrent aspiration pneumonitis, special procedures like thyroplasty and cricopharyngeal myotomy may be advocated., A less-often discussed morbidity associated with the radical excision of these lesions is the occurrence of alteration of sensorium, axial ataxia, titubation and long tract signs due to brain stem handling, spreading venous thrombosis and perforator compromise during excision of these tumors. Leaving a thin rim of tumor tissue attached to the brain stem and then serially and meticulously following up these patients to detect any signs of growth, significantly, reduces the occurring of these disabling features in patients with petroclival meningiomas.,,,,
To conclude, an aggressive surgical resection has a predominant role in alleviating the symptomatology arising due to the presence of extensive petroclival meningiomas. The middle fossa approaches help in significantly avoiding morbidity as they enable an early devascularisation and decompression of the tumor. The temptation to radically remove an extensive petroclival meningioma that lacks a plane of cleavage from the brain stem should be resisted.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7], [Figure 8], [Figure 9], [Figure 10]
[Table 1], [Table 2], [Table 3], [Table 4]