527 Fully Endoscopic Resections of Vestibular Schwannomas
By H. K. Shahinian , Y. Ra
We report a series of 527 patients with unilateral vestibular schwannomas (VS) who underwent fully endoscopic resection of their tumors during the period of October, 2001 to July, 2010. Patients ’ outcomes were evaluated, with specific regard to hearing preservation, facial nerve function, postoperative complications and completeness of the resection.
The patient population consisted of 527 cases with unilateral VS(s); patients with neurofibromatosis type 2 (NFT2) were excluded from this study. Tumors ranged in size from 0.3 - 5.8 cm, most tumors were less than 4 cm in diameter (mean: 2.8 cm). Tumors were removed via 2.0 cm "keyhole" retrosigmoid craniotomies.
Utilizing the fully endoscopic technique, 94% of tumors were completely removed; subtotal removal was performed in 6% of patients in an attempt to preserve their hearing. Anatomic preservation of the facial nerve was achieved in all of the patients. Functionally, measurable hearing (serviceable / some) was preserved in 57% of cases that had either "serviceable" or "some" hearing pre-operatively. There were no major neurological complications such as quadriparesis, hemiparesis, bacterial or aseptic meningitis, permanent lower cranial nerve deficits, or deaths.
From our experience, we conclude that the endoscope is ideally suited for a minimally invasive approach for resection of vestibular schwannomas.
Endoscopic, Acoustic Neuroma, Vestibular Schwannoma
Vestibular schwannoma (VS), also known as acoustic neuroma or neurinoma, is a benign overproliferation of the Schwann cells of the eighth cranial nerve sheath, usually starting at the junction between the peripheral and the central myelin sheath  . The estimated incidence of VS(s) based on cadaveric dissections by Schuknecht is said to be 570 / 100 000  . However, many of these tumors fail to become symptomatic during a patient's lifetime. A consensus statement by the National Institutes of Health in 1991 estimated that between 2 - 3 000 vestibular schwannomas are diagnosed each year in the United States of America, representing a symptomatic incidence of only 1 / 100 000. VS(s) are reported to occur in all races  ; they account for 6 - 10% of all primary intracranial tumors [1, 4, 5] and about 71 - 90% of all cerebellopontine angle tumors. Most commonly, VS(s) occur during the fourth and fifth decades of life  and are about 2 times as common in females as in males  . There are 2 distinct clinical presentations for VS(s): sporadic, unilateral VS and hereditary, bilateral VS. Bilateral schwannoma of the internal auditory canal (IAC) is often an expression of neurofibromatosis type 2 (NFT2)  . Symptoms and signs associated with VS(s) have been known for over 150 years and are due to direct tumor compression, invasion or vascular compromise of the surrounding structures  . VS(s) are commonly associated with gradual (over months or years) hearing loss indicating damage to the cochlear nerve  . They may manifest with other symptoms and signs depending on the site of origin and the growth pattern of the tumor; including tinnitus, vertigo, cerebellar dysfunction, cranial nerve dysfunction, and secondary obstructive hydrocephalus  .
The first successful removal of an acoustic tumor was accomplished by either Sir Charles Ballance in 1891 or Thomas Annandale in 1895  . At the beginning of the last century, surgeons were content when they merely managed to save the life of the VS patient. During the last few decades, the advent of microsurgery, advanced imaging technology, and intra-operative electrophysiological monitoring has shifted the focus to facial nerve preservation and preservation of serviceable hearing; and to the target of gross total tumor removal.
Several reports have highlighted the utility of the endoscope to assist microscopic removal of acoustic neuromas [8, 9] . These articles suggest that the endoscope provides improved recognition of exposed air cells and allows for more complete tumor removal by direct visualization of the IAC to remove any residual tumor out of the view of the operating microscope [8, 9] .
Our group has been performing endoscope-assisted and fully endoscopic surgery of the CPA for trigeminal neuralgia, hemifacial spasm, glossopharyngeal neuralgia and CPA tumors since 1996 [10 - 13] . We began using the endoscope to supplement our microsurgical resection of VS(s) in l998. Using the endoscope in this region, we found that the maneuverability and angled lenses of the endoscope provide significant advantages in visualizing and accessing the entire tumor, while avoiding injury to the surrounding neurovascular structures and eliminating the need for any cerebellar retraction. Late in 2001, our group converted to a fully endoscopic technique for resection of VS(s) [14, 15] . In this report we present our experience with 527 cases where the endoscope was used as the sole imaging modality in performing VS surgery.
Patients and Methods
The subject population consists of 527 patients with unilateral VS who have undergone fully endoscopic surgical resection of their tumors from October, 2001 to July, 2010. Patients with NFT2 were excluded from this study. This series included 44 cases that had undergone previous gamma knife radiosurgery and 61 cases with purely intracanalicular VS(s). Assessment was based upon clinical, radiological, and audiometric examinations. Pre- and postoperative data were collected and evaluated in a database for patients who have undergone fully endoscopic VS resections at the Skull Base Institute in Los Angeles, California. Demographic information is presented in Table 1 .
All patients were prepared for surgery by thorough clinical investigation, including audiometry, brainstem auditory evoked response (BAER), contrast-enhanced CT with bone window, and gadolinium-enhanced MRI. Postoperatively; clinical, and contrast-enhanced MRI follow-up examinations were scheduled at 3 - 6 months, 1 year, and then annually for the rest of the followup period. The completeness of tumor removal was judged by surgical records (intraoperative digital pictures) and postoperative MRI.
Outcomes were evaluated especially with regards to cochlear (hearing) and facial nerve preservation, postoperative complications, and the completeness of resection. Audiometric parameters were defi ned by the terms " serviceable " hearing, " some" hearing and " total deafness "  .
The operation begins with the patient placed in a lateral "parkbench" position; the patient's head is secured in a Mayfield 3-pin head clamp. The head is then flexed and slightly rotated away from the side of the tumor.
A 2.5 cm retroauricular incision is performed. Using the Asterion as a bony anatomic landmark, a 2.0 cm craniotomy is made at the confluence of the sigmoid and transverse sinuses using a high-speed drill. The bone flap is removed in one piece to be repositioned at the end of the operation ( Fig. 1 ) . Bone wax is used to fill any mastoid air cells entered during the bone drilling. An incision is made in the dura, which is retracted anteriorly. CSF is allowed to slowly drain and a combination of mild hyperventilation, mannitol and positioning allows the cerebellum to spontaneously retract, opening up a narrow path to the cerebellopontine angle. A 2.7 or 4 mm zero-degree endoscope (Storz, Culver City, CA) is then guided atraumatically along this path with minimal dissection and almost no retraction to visualize the tumor. An irrigation sheath attached to the endoscope clears blood and debris from the lens, eliminating the time-consuming and unsafe practice of removing and re-inserting the endoscope. A rigid pneumatic holding arm secures the endoscope in position, allowing bimanual surgical dissection.
Upon entering the CPA, the surgeon conducts a preliminary survey of the surrounding structures including the trigeminal, facial, and lower cranial nerves, as well as the regional vascular anatomy. The facial nerve is protected via a facial nerve monitor (Xomed, Jacksonville, FL), which remains in place for the duration of the operation to avoid injury to the facial nerve. Once the surrounding critical structures are identified, tumor dissection takes place guided by a zero-degree endoscope in much the same manner as the microsurgical procedure with custom designed microinstruments which slide along the endoscope. Using a combination of the micro-cavitron ultrasonic surgical aspirator (microCUSA), bipolar electrocoagulation, suction-irrigation and microsurgical instruments customized for this endoscopic approach, the interior of the tumor is excised first. The capsule of the tumor is then gently dissected off the brainstem medially and facial nerve anteriorly. A diamond burr is then used to remove the posterior wall of the internal auditory canal, thus following the tumor laterally within the canal. In cases of smaller tumors with patients having "serviceable" hearing preoperatively, this portion of the dissection should be performed with extreme caution. Following entry into the IAC, the zero degree endoscope is removed and the thirty-degree endoscope is introduced. Tumor dissection within the IAC is guided by the angled endoscope, allowing complete visualization of the lateral extent of the tumor as it is separated from the facial nerve
( Fig. 2 ).Once tumor dissection is complete, the facial nerve is stimulated to confirm its function. The dura is re-approximated; the bone flap replaced and secured with a resorbable plate, and the scalp is closed in anatomic layers without the use of any drains. Following the operation, patients are typically transferred to a private room for overnight monitoring and then discharged within 48 h postoperatively.
Intraoperative pictures presenting 5 cases of fully endoscopic resection of VS tumors of different sizes are given. The images present the initial visualization of the tumor, half-way through debulking of the tumor and after complete resection ( Fig. 3 ).Pre- and postoperative gadolinium-enhanced MRIs of patients with vestibular schwannomas of diff erent sizes that were resected by the fully endoscopic technique ( Fig. 4 ).
In this study of 527 unilateral cases of VS, the size of the tumors ranged from 0. 3 - 5.8 cm in diameter, the vast majority being less than 4 cm in diameter (mean: 2.8 cm). Total excision was the primary aim in all cases and was given the highest priority. Recurrences occurred in 12 out of 527 patients. The presenting clinical manifestations of the patients are shown in Table 2 . The most common presentations were hearing deterioration or loss in 469/ 527 (89%); gait instability in 353/ 527 (67%); persistent headaches in 316 / 527 (60% ); tinnitus in 274 / 527 (52%); trigeminal paresthesia or hypesthesia in 116 / 527 (22% ).
Facial nerve function
Facial nerve function was evaluated using House-Brackmann (H-B) grades 1 - 6 , and was categorized as excellent (H-B grade 1 / 2), intermediate (H-B grade 3 / 4), or poor (H-B grade 5 / 6). Anatomic preservation of the facial nerve was achieved in 527 / 527 (100%) cases. One year after tumor removal, facial nerve function was re-assessed. Out of 527 patients with anatomically preserved facial nerves, 491 / 527 (93 % ) showed excellent facial nerve function (H-B grade 1 / 2); while 21 / 527 (4 % ) showed intermediate function (H-B grade 3 / 4); and 15 / 527 (3% ) showed poor function (H-B grade 5 / 6).
Cochlear nerve (Hearing) preservation
The audiometric parameters are defined as follows: "serviceable" hearing - less than 50 dB hearing loss; "some" hearing - 50- 80 dB hearing loss; and "total deafness" - greater than 80 dB hearing loss. Functional hearing preservation, being defined as measurable hearing (serviceable / some), was possible in 213 / 374 (57%) of the same group. One patient with a right-sided intracanalicular VS, 10 mm in diameter and with a small portion of it protruding into the CPA, initially had "some" hearing and regained a "serviceable" hearing one week postoperatively. Another patient had an improvement of > 20 dB in his hearing postoperatively ( Table 3 ).
Postoperative complications unrelated to cranial nerves are shown in Table 4 , as the most relevant ones, related to cranial nerves, are separately discussed. CSF leak from the wound occurred in 17 cases, either re-suturing of the wound or bone waxing air cells stopped the leak (lumbar drain was not required) in all but one patient who required a subtotal petrosectomy.
One patient developed postoperative ventricular dilatation on the second day of surgery and was obtunded; a temporary ventricular catheter for diversion of CSF was placed for 5 consecutive days and the condition resolved. Exposure keratitis occurred in 9 patients and was treated with a "gold-weight" placed over the affected eyelid in 3 patients and aggressive ointment and artificial tears in the other 6 patients until facial nerve function eventually improved. Superficial wound infection was encountered in 13 patients and was treated conservatively in 4 patients; the other 9 patients required re-opening of the wound for irrigation, disinfection and overall debridement.
The diagnosis of VS has changed considerably during the last years as magnetic resonance imaging (MRI) has become the preferred diagnostic tool  . In our series, we have noticed a shift towards smaller tumors with minor symptoms accompanied by a corresponding increase in patient awareness. The advent of specialized MRI / MRA imaging and the development of endoscopic surgical instruments and techniques along with smaller tumor sizes have definitely influenced the postoperative results. Since major complications such as hemorrhage, meningitis, brainstem compression injuries and cerebrospinal fluid (CSF) leak are becoming less common, quality of life issues such as disequilibrium, headache, and hearing preservation are becoming more important. Furthermore, there is a general agreement that completeness of resection and preservation of the facial nerve are the major goals  and they are successfully being met at increasing rates.
The pioneering effort of William House has paved the way for renewed optimism in the successful management of VS(s) with a low mortality and acceptable morbidity. Drake, Rand and Kurze, and others showed that acoustic tumors could be removed totally, safely, and with preservation of the facial nerve. Rand and Kurze discussed the possibility of anatomic preservation of the cochlear nerve, which they subsequently demonstrated in 1968 [19 - 21] .
The retrosigmoid approach to the posterior fossa, first described by Cohen  in 1992, is actually a modification of the suboccipital craniotomy or craniectiomy. Its advantages include that it gives a wide view of the posterior fossa and offers a good chance of cranial nerve preservation. The fully endoscopic approach is based on the very same principles of this approach. However, disadvantages of the operating microscope with its direct forward view or inability to "look around the corner" disallows the surgeon to completely visualize the lateral extent of the tumor within the IAC [8, 9] , as well as exposed air cells, which may lead to CSF rhinorrhea  .
We have found that the fully endoscopic approach allows for the possibility of hearing preservation, provides excellent visualization of the entire tumor, avoids blind dissection behind the facial nerve, and is well tolerated by patients with minimal discomfort. We believe that the endoscopic approach allows for smaller craniotomies, requires less dissection, and virtually, no cerebellar retraction.
Facial nerve preservation
In a study by Hardy et al., in 1989, using a translabyrinthine (TL) approach, they were able to preserve the facial nerve in 82 % of the 98 patients with intact nerve preoperatively. Postoperatively, 77 patients assessed their life quality to be excellent  .
In this series, at the time of discharge from the hospital, most of the patients had satisfactory facial nerve function with complete eye closure. After one year, 491 / 527 (93% ) showed excellent facial nerve function (H-B grade 1 / 2), and 21 / 527 (4% ) showed intermediate function (H-B grade 3 / 4) [patients operated after July 2009 did not have a full year follow-up to assess their facial nerve function].
In spite of the high rate of facial nerve preservation in this study, anatomic preservation of the facial nerve with complete tumor removal, especially in patients with large tumors, is still a challenge. Facial nerve monitoring has greatly aided separation of the facial nerve from the tumor [8, 9, 24 - 26] .
Cochlear nerve (Hearing) preservation
Cochlear nerve preservation has been reported by many authors [10, 11, 18, 29, 30] in the recent past, but there is a lot of ambiguity on the criteria for useful hearing. In addition, there are some who believe that the goal of gross total tumor removal cannot be achieved with cochlear nerve preservation  .
In this report, our rate of functional hearing preservation (213 / 374) (57% ) reflects the better outcome associated with better visualization by the angled and zero-degree high-definition endoscopes. The fact that most tumors in this series were less than 4 cm in diameter and presented early with minimal symptoms also has a direct impact on the surgical outcome and prognosis.
Completeness of resection
In the majority of cases a VS originates from the vestibular nerve and only compresses the cochlear nerve  . More than 95 % of VS(s)  arise from vestibular fascicles; Samii et al. in their series of 1 000 acoustic neuromas have observed that only 1.1 % of the CPA tumors arise from the cochlear nerve  . Therefore, resection of a macroscopically intact cochlear nerve in an attempt to seek complete tumor removal is not advised by many authors.
In our series, by opening the IAC under direct visualization of zero-degree and angled endoscopes, and by gradual reduction of the tumor as long as brainstem auditory evoked potentials were stable and reproducible, total tumor removal was achieved with hearing preservation. In these cases, tumor remnants were not left behind for the sake of hearing preservation, but rather, were followed and carefully excised. Subtotal tumor removal in this study was done for 31 / 527 (6 % ) patients and was performed only when a "tumorous" cochlear nerve was encountered in a patient who had "serviceable" or "some" hearing. In these cases, the benefit of total removal was and should be carefully weighed against the risk of losing a functioning cochlear nerve.
The importance of complete tumor removal and the effect it has on recurrence is well known. In 1989, Hardy, et al. reported only 3 perioperative deaths among 100 TL VS(s) operations, and the postoperative morbidity was low  . Complete tumor excision was achieved in 97 % of cases and no recurrences were seen during follow-up of 1 - 7 years. In that series, hearing preservation was of no concern.
Cerebellar and brain stem injuries are the major and most feared complications of the retrosigmoid approach [26, 29] . Postoperative complications excluding those related to cranial nerves were minimal ( Table 4 ). Major complications such as postoperative hemorrhage, quadriparesis, hemiparesis, meningitis or death did not occur. As the fully endoscopic retrosigmoid approach does not require the use of metal retractors to manipulate brain tissue for access and visualization of the tumor, complications are generally less frequent when compared to traditional open craniotomies. Bacterial or aseptic meningitis also occurs less frequently because the operative field is reduced, along with the duration of surgery and hospitalization.
Hemorrhage into the posterior fossa or subdural or extradural hemorrhage in the immediate postoperative period can produce brainstem compression and rapid death. The prevention of secondary bleeding relies on strict homeostasis, and it is essential to coagulate both ends of any minor surface vessels during tumor resection. A custom designed microbipolar electrocautery system allows for meticulous hemostasis.
Early or delayed CSF leakage from the wound, rhinorrhea, or otorrhea (either directly through the wound or via the Eustachian tube and middle ear) can occur even if all the precautions have been taken to avoid them. In some cases, the leakage may be iatrogenic; this emphasizes the importance of watertight dural closure. The path of CSF egress is most commonly through mastoid air cells that were opened during the craniotomy. The keyhole craniotomy, even though limited, can still result in some postoperative headaches, but these resolve spontaneously in the majority of cases. Taking great care to avoid contaminating CSF with bone dust and replacement of the keyhole bone flap have significantly reduced the incidence of this complication in our experience. Mild transient postoperative hydrocephalus rarely occurs in the early postoperative period; even when present, it generally resolves without difficulty in the first few postoperative weeks, and CSF shunting is rarely, if ever, required.
Improved diagnostic screening with MRI and a better informed population have resulted in the diagnosis of smaller and even asymptomatic VS(s) [2, 30] . The endoscope provides improved visualization of the skull base, where narrow recesses and angled trajectories impair the direct forward view of the operating microscope. Endoscopic surgery allows for smaller craniotomies, less dissection and minimal retraction, without compromising the goals of the operation. 2 important factors with regards to predicting the preservation of cranial nerves VII and VIII are tumor size and preoperative hearing status.
We believe that the improved exposure of the entire tumor provided by the endoscope with minimal or no retraction reduces the risk of injury to the brainstem and the surrounding cranial nerves, and results in a complete tumor removal. The more direct "keyhole" approach has significantly decreased the time required for exposure of the tumor and the overall operative time (193 min) in this series. This minimally invasive technique allowed rapid recovery of the patients (mean LOS: 2.4 days) and resulted in minimal postoperative discomfort [14, 15] .
Conflict of Interest:None
- Shiff man F, Dancer J, Rothballer AB et al . The diagnosis and evaluation of acoustic neuromas . Otolaryngol Clin North Am 1973 ; 6 : 189 - 228
- Rosenberg SI. Natural history of acoustic neuromas . Laryngoscope 2000 ; 110 : 497 - 508
- Lanser MJ, Sussman SA, Frazer K . Epidemiology, pathogenesis, and genetics of acoustic tumors . Otolaryngol Clin North Am 1992 ; 25 : 499 - 520
- Nager GT. Acoustic neurinomas . Acta Otolaryngol (Stockh) 1985 ; 99 :245 - 261
- Tos M, Charabi S, Thomsen J . Clinical experience with vestibular schwannomas: epidemiology, symptomatology, diagnosis, and surgical results . Eur Arch Otorhinolaryngol 1998 ; 255 : 1 - 6
- Selesnick SH, Jackler RK . Clinical manifestations and audiologic diagnosis of acoustic neuromas . Otolaryngol Clin North Am 1992 ; 25 : 521 - 551
- Cushing H. Tumors of the nervus acusticus and the syndrome of the cerebellopontine angle . W.B. Saunders, Philadelphia/London ; 1917
- Low WK. Enhancing hearing preservation in endoscopic-assisted excision of acoustic neuroma via the retrosigmoid approach . J Laryngol Otol 1999 ; 113 : 973 - 977
- Magnan J, Barbieri M, Mora R et al . Retrosigmoid approach for small and medium-sized acoustic neuromas . Otol Neurotol 2002 ; 23 : 141 - 145
- Eby JBC, Shahinian HK. Fully endoscopic vascular decompression of the facial nerve for hemifacial spasm . Skull Base: An Interdisciplinary Approach 2001 ; 11 : 189 - 196
- Jarrahy R, Berci G, Shahinian HK . Endoscope-assisted microvascular decompression of the trigeminal nerve . Otolaryngol Head Neck Surg 2000 ; 123 : 218 - 223
- Jarrahy R, Cha ST, Berci G et al . Fully endoscopic vascular decompression of the glossopharyngeal nerve . J Craniofacial Surg 2002 ; 13 : 90 - 95
- Jarrahy R, E by JB, C ha ST et a l . Fully endoscopic vascular decompression of the trigeminal nerve . Minim Invas Neurosurg 2002 ; 45 : 32 - 35
- Shahinian HK, Eby JB, Ocon M . Fully endoscopic excision of vestibular schwannomas . Minim Invas Neurosurg 2004 ; 47 : 329 - 332
- Kabil MS, Shahinian HK. A series of 112 fully endoscopic resections of vestibular schwannomas. M inim Invas Neurosurg 2 006; 4 9: 3 62- 3 68
- House J, Brachman DE. Facial nerve grading system . Otolaryngol Head Neck Surg 1985 ; 93 : 146 - 147
- Robinette MS, B auch C D, O lsen WO e t a l. Auditory brainstem response and magnetic resonance imaging for acoustic neuromas . Arch Otolaryngol Head Neck Surg 2000 ; 126 : 963 - 966
- Ramsay HA, Luxford WM. Treatment of acoustic tumours in elderly patients: Is surgery warranted? J Laryngol Otol 1993 ; 107 : 295 - 297
- House WF, Monograph I. Transtemporal bone microsurgical removal of acoustic neurinomas . Arch Otolaryngol 1964 ; 80 : 597 - 756
- Drake CG. Total removal of large acoustic neuromas . J Neurosurg 1967 ; 26 : 554 - 561
- Rand R, Kurze T. Preservation of vestibular, cochlear and facial nerves during microsurgical removal of acoustic tumors: Report of two cases . J Neurosurg 1968 ; 28 : 158 - 161
- Cohen NL. Retrosigmoid approach for acoustic tumor removal . Otolaryngol Clin North Am 1992 ; 25 : 295 - 310
- Hardy DG, MacFarlane R, Baguley D et al . Surgery for acoustic neuroma.An analysis of 100 translabyrinthine operations . J Neurosurg 1989 ;71 : 799 - 804
- Axon PR, Ramsden RT. Facial nerve injury caused by vestibular schwannoma compression: severity and adaptation to maintain normal clinical facial function . Am J Otol 1999 ; 20 : 763 - 769
- Post KD, Eisenberg M, Catalano P . Hearing preservation in vestibular schwannoma surgery: what factors infl uence the outcome? J Neurosurg 1995 ; 83 : 191 - 196
- Samii M, Matthies C. Management of 1 000 vestibular schwannomas: surgical management and results . Neurosurgery 1997 ; 40 : 11 - 23
- Tos M, Thomsen J, Harmsen A. Is preservation of hearing acoustic neuroma worthwhile? Acta Otolaryngol (Stockh) 1988 ; 452 : 57 - 68
- Horrax G, Poppen JL. The end results of complete versus intracapsular removal of acoustic tumors . Ann Surg 1949 ; 130 : 567 - 575
- Wiet RJ, Teixido M, Liang JG. Complications in acoustic neuroma surgery. Otolaryngol Clin North Am 1992 ; 125 : 389 - 412
- Selesnick SH, Deora M, Drotman MB et al . Incidental discovery of acoustic neuromas . Otolaryngol Head Neck Surg 1999 ; 120 : 815 - 818