- Email: [email protected]
Surgery combined with radiosurgery of large acoustic neuromas
Neoplasm
Surgery Combined With Radiosurgery of Large Acoustic Neuromas Yoshiyasu Iwai, M.D., Kazuhiro Yamanaka, M.D., and Tomoya Ishiguro, M.D. Department of Neurosurgery, Osaka City General Hospital, Osaka, Japan
Iwai Y, Yamanaka K, Ishiguro T. Surgery combined with radiosurgery of large acoustic neuromas. Surg Neurol 2003;59:283–91. BACKGROUND
The treatment of acoustic neuromas has been improved by advancements in microsurgical techniques and in radiosurgery. To further elucidate the degree of clinical improvement, we evaluated the treatment results of a combination of surgery and radiosurgery for large acoustic neuromas. METHODS
From January 1994 through December 2000, we treated 14 patients with large acoustic neuromas using a combination of surgery and radiosurgery. Of these, 8 were male and 6 were female patients, with an average age of 47 years (range, 18 – 64). The average maximum diameter of the tumor was 42 mm (range, 30 –58 mm). All patients underwent operations using the retrosigmoid approach, and one patient was retreated using the transpetrosal transtentorial approach. The tumors were removed subtotally in thirteen patients and partially in one who had a very large hypervascular acoustic neuroma. There were no mortality and no surgical complications, such as hemorrhage or CSF leakage. Postoperative facial palsy was avoided in 10 patients (71%). Radiosurgery was performed 1 to 6 months (mean, 2.9 months) after surgery. At the time of radiosurgery, the treatment size (mean diameter) became 19.2 mm (range, 9.8 –36.1 mm). The average tumor marginal dose was 12.1Gy (range, from 10 –14 Gy). The mean follow-up period was 32 months after radiosurgery. RESULTS
The tumor size decreased in 6 patients, unchanged in 5 patients, and increased in 3 patients. Only 1 patient (7%) with extra large tumor needed surgical resection 1 year after radiosurgery. Excellent facial nerve function (House & Brackmann Grade I or II) was preserved in 12 patients (85.7%) in the final follow-up. CONCLUSIONS
In the case of large acoustic neuromas, subtotal removal and subsequent radiosurgery is one option for maintaining cranial nerve function and long-term tumor growth control. © 2003 Elsevier Inc. All rights reserved. Address reprint requests to: Dr Yoshiyasu Iwai, Department of Neurosurgery, Osaka City General Hospital, 2-13-22, Miyakojima-hondohri, Miyakojima-ku, Osaka, 534-0021, Japan. Received April 26, 2002; accepted November 20, 2002. © 2003 Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010 –1710
KEY WORDS
Vestibular schwannoma, surgery, radiosurgery, acoustic neuroma.
or the treatment of acoustic neuromas, excellent surgical results have been reported recently [27]. However, in terms of functional outcome, the surgical results for large acoustic neuromas may be worse compared to small or medium-sized tumors. Therefore, for large acoustic neuromas, intracapsular removal is one of the techniques used by some surgeons to preserve cranial nerve function [18]. However, the cases of subtotal removal have revealed a high recurrence rate [1]. We innovated the idea of intracapsular removal, thus allowing subtotal removal combined with radiosurgery, to increase the odds of preserving cranial nerve function and maintaining long-term tumor growth control.
F
Materials and Methods In our institute since January 1994, with a gamma knife unit (Elekta Instruments, Atlanta, Georgia, AL), we have taken maximum advantage of radiosurgery in the treatment of acoustic neuromas. We treated our series of acoustic neuromas in the following manner: The small- to medium-sized tumors with extrameatal diameters less than 3 cm were treated by radiosurgery, while the large tumors (defined as lesions with maximum extrameatal diameters greater than 3 cm) with neurologic symptoms, such as cerebellar ataxia and trigeminal neuropathy and other symptoms besides hearing disturbances, were treated by a combination of surgery and radiosurgery. Because these large tumors tend to cause clinical symptoms in addition to hearing disturbances, they must be resected to improve this symptomatology. 0090-3019/03/$–see front matter doi:10.1016/S0090-3019(03)00025-9
284 Surg Neurol 2003;59:283–91
Iwai et al
Preoperative Cranial Nerve Deficits and Clinical Signs Noted in 14 patients with Large Acoustic Neuromas
1
CLINICAL SIGNS 5th nerve 8th nerve Cerebellar sign Papilledema
NO.
%
7 14 8 4
50 100 57 29
According to this treatment strategy, we operated on 17 patients and 14 of these patients underwent radiosurgery from January 1994 through December 2000. In that period, 3 surgery patients were followed without radiosurgery. Of these, 2 patients received radiosurgery thereafter, but these 2 patients were not included in this study because only short follow-ups were possible after radiosurgery. One patient was observed after surgical removal without subsequent radiosurgery in accordance with the patient’s request. In this patient, the residual tumor was unchanged during 3 years after operation. Thus, we evaluated 14 patients who had undergone combined treatment (operation and radiosurgery). The study population was composed of 8 women and 6 men. The age range was 18 to 64 years (mean: 47 years). Two patients suffered from NF2. All patients suffered complications other than hearing disturbances, such as fifth cranial nerve palsy in 7 patients (50%); cerebellar ataxia in 8 patients (57%); and papilledema in 4 patients (29%) (Table 1). The hearing disturbance was evaluated according to the Gardner and Robertson (G&R) classification’s system [3]. There was one Class 1 patient, two Class 2 patients, two Class 3 patients, five Class 4 patients, and four Class 5 patients. Useful hearing (G&R class 1 or 2) was preserved in only 3 patients (21%). The maximum tumor diameter at the extrameatal portion was 30 to 40 mm in 6 patients, 40 to 50 mm in 6 patients, and above 50 mm in 2 patients (Table 2). The operation was performed via a retrosigmoid approach in all patients. In 1 patient with a very large hypervascular tumor associated with NF2, the
2
The Size Distribution of 14 Patients with Large Acoustic Neuromas
TUMOR SIZE (MM) 30–40 40–50 50⬍
NO.
%
6 6 2
43 43 14
tumor could not be removed by this approach because of cerebellar edema; thus, the transpetrosal transtentorial approach was employed. The operation was performed in the supine position. The tumor was removed in the following manner: The CSF was released from the cerebellopontine angle cistern. The arachnoid covering the tumor was incised and dissected. The dorsal part of the tumor was stimulated by a nerve stimulator to ensure that the facial nerve was not running in or near the dorsal portion of the tumor. The internal decompression of the tumor was performed by using the ultrasound aspiration and suction. The resection of the tumor capsule was performed from the cerebellar surface to the junction of the cerebellum and brainstem. The tumor around the internal auditory meatus and anterior portion close to the brain stem was left in place because manipulation of this part of the tumor could injure the facial nerve. The portion of tumor close to the lower cranial nerve was left in cases where the tumor dissection from these nerves was difficult. After tumor removal, the dura was sutured tightly and the suture site was sealed by fibrin-soaked Gelfoam to prevent postoperative CSF leakage, and the bone was replaced and fixed by a titanium plate. The interval between surgery and radiosurgery was 1 to 6 months (mean: 2.9 months). While the tumor seemed to adhere closely to the brainstem on image immediately after operation, after 3 months the tumor became a more appropriate shape for radiosurgery (Figure 1). In 2 patients with a cystic tumor, radiosurgery was performed 1 month after operation to prevent the growth of the cystic part of the tumor. In 1 patient with a very large hypervascular tumor (mean diameter: 58 mm), the radiosurgery was performed in a twostaged procedure with a 3-month interval between the stages, as we use for large skull base meningiomas [11,24]. The radiosurgery was performed for treatment size having diameters of 9.8 to 36.1 mm (mean: 18.9 mm) with tumor marginal doses ranging from 10 to 14.1Gy (mean: 12.1Gy). The patients could return to normal activity immediately after radiosurgery.
Results The tumor was removed subtotally in thirteen patients (93%) and partially in one patient (7%). The patient with partial removal had a very large hypervascular tumor, and we could not remove the tumor radically because of intraoperative bleeding. We did not experience operative mortality or other
Treatment for Large Acoustic Neuromas
Surg Neurol 285 2003;59:283–91
Tumor change after operation. A: Preoperative magnetic resonance imaging with gadolinium-enhancement showing a large acoustic neuroma at the left cerebellopontine angle in a 27-year-old woman. B: One month after operation, the residual tumor was situated at the left cerebellopontine angle. C: Three months after operation, the tumor shape became suitable for radiosurgery.
1
complications including CSF leakage or hemorrhage. One patient suffered trochlear nerve palsy. Facial palsy occurred in 4 patients (29%). The cause of the facial palsy cases was considered to be the
extensive manipulation of the ventral site of the tumor capsule, and the facial palsies were complete immediately after the operations. However, in the final follow-up, the facial palsy was improved to
286 Surg Neurol 2003;59:283–91
House and Brackmann (H&B) Grade II in 2 patients and Grade III in 2 patients [9]. We did not suffer other neurologic deficits after operations. Preexisting trigeminal neuropathy, cerebellar ataxia, and increasing intracranial pressure resolved in all patients during the early stages after the operations. Among 3 patients with useful hearing (with G&R Class 1 and 2 preoperatively), postoperatively useful hearing was preserved in only 1 patient. We were able to follow all patients in this study for 12 to 72 months (mean: 32 months) after radiosurgery. The tumors decreased in size in 6 patients (43%), remained unchanged in 5 patients (36%), and increased in size in 3 patients (21%) (Figure 2). Only 1 patient (7%) deteriorated because of tumor growth. In 1 patient with tumor growth, the tumor increased in size gradually during the 5 years after radiosurgery, and now remains stable in size. In 1 patient with NF2, the tumor was slightly increased in size during 24 months after radiosurgery. In the other patient (who had a very large hypervascular tumor with NF2), we could not remove the tumor radically, and this tumor with a mean diameter of 58 mm was treated by a two-staged radiosurgery procedure. In this patient, the tumor had been increasing in size and causing deterioration in neurologic condition; it was removed 1 year after radiosurgery. Interestingly, the tumor became avascular on angiographic imaging and was able to be removed subtotally. No other patient deteriorated after radiosurgery. No patient suffered radiation injury including facial palsy.
Discussion Lownie et al reviewed the historical surgical results of acoustic neuromas [18]. From their article, Cushing suffered significant mortality when attempting complete removal of the tumor, and he advocated intracapsular removal of the tumor. However, it was later demonstrated that such partial removals have a high recurrence rate [6]. Olivecrona et al reported good results of complete removal of the tumor in some patients and high rate of tumor regrowth after incomplete removal. But they still had high mortality for these procedures [4,22]. Mckenzie et al reported excellent results of intracapsular removal with low incidence of lower cranial nerve dysfunction, brain stem injury, and facial palsy [20]. Thus, microsurgical resection has been the primary treatment of acoustic neuromas since the microscope’s first use for acoustic neuroma surgery in 1961 [7]. Microsurgical resections allow radical tumor resections with less morbidity and mortality,
Iwai et al
but for large tumors the procedure still resulted in a high rate of facial palsy (37– 80%) in the 1980s [5,8,26,28]. Recently, excellent treatment results of surgery for acoustic neuromas have been reported. Samii reported 1,000 cases of surgery for acoustic neuromas and that his group could achieve total removals in 98% of cases. In addition, he noted facial nerve preservation in 87% of the patients [26]. Lamnan et al treated 190 patients who had large acoustic neuromas among 900 patients operated during the same period, 1990 through 1995, and achieved a total removal percentage of 96%. Low morbidity and excellent facial nerve function (H&B Grade I or II) was preserved (52.6%) in the patients on the final follow-up [16]. In the patients with extra large tumors (maximum extrameatal diameter above 4 cm), Jung et al reported their excellent surgical results with 73.3% of total removal and 86.4% of facial nerve preservation, and excellent facial nerve function (H & B Grade I or II) was preserved in 42.1% of patients [12]. So, the preservation of facial nerve function for the large acoustic neuromas was still difficult with radical removal of the tumors. Malis also commented that a surgeon needed the 200 cases to achieve excellent results with acoustic neuromas [19]. Lownie et al reported intracapsular removal of large acoustic neuromas in 11 patients with long follow-up for functional preservation during the era of microsurgery. The facial nerve was preserved in 9 patients (82%), but in 2 patients (18%) the residual tumor increased in size 2 and 3 years after surgery [18]. Kameyama et al followed 19 patients with residual tumor after intracapsular removal and 10 patients (53%) had regrowth. Among them, 5 patients with cyst formation had rapid regrowth (median, 4.5 years) and required reoperation, but the solid tumors were growing slowly (median, 15 years) [13]. Ohta et al reported that the tumor leaving the site of internal auditory canal had increased in size, and the tumor exiting the brainstem (and before entrance of internal auditory canal) had not increased during the follow-up period of 4.5 to 8 years [21]. El-Kashlan et al followed 39 patients with residual tumor for a mean period of 6.2 years, and the residual tumor was increased in size in 17 patients (44%) [1]. So, to reduce operative morbidity and preserve the functional outcome, intracapsular removal may be one operative approach in the era of microsurgery. But from the standpoint of tumor growth control, some adjuvant therapy may be needed for long-term tumor growth control. Radiosurgery has a long history of treating acous-
Treatment for Large Acoustic Neuromas
Surg Neurol 287 2003;59:283–91
A: Axial view of magnetic resonance (MR) imaging with gadolinium-enhancement showing a large tumor at the left cerebello-pontine angle compressing the brain stem of a 54-year-old female with clinical symptoms of left-sided deafness and cerebellar ataxia. B: The residual tumor after subtotal removal via the retrosigmoid approach without any neurologic deterioration. C: The radiosurgical treatment with a dose to the periphery of the tumor of 12 Gy. D: Four years after radiosurgery, MR imaging showing the tumor not having increased in size and the patient in good health with no new neurologic deficits.
2
tic neuromas. The first acoustic neuroma patient was treated in 1969 at Kalorinshka Hospital, using a gamma knife unit [17]. The treatment results were
improved by reducing the radiation dose and obtaining higher-quality images [2,14]. Konziolka et al reported the 5-year follow-up results in 165 patients
288 Surg Neurol 2003;59:283–91
and achieved 98% clinical tumor growth control (without tumor resection) [14]. Flickinger et al reported from the same institute as Konziolka that their tumor growth control rate was 91.0 ⫾ 2.5%. Few patients suffered facial palsy (1.1 ⫾ 0.8%) and 0% of those who received less that 13 Gy for tumor margin suffered facial palsy. Serviceable hearing was preserved in 73.5 ⫾ 4.7% of patients after 5 years [2]. Prasad and Steiner treated 159 patients with acoustic neuromas via gamma knife radiosurgery; they reported 89% tumor growth control in patients that had undergone previous operations and 94% tumor growth control in the nontreated patients along with 40% hearing preservation. Facial palsy occurred in only 2% and new trigeminal neuropathy occurred in 3% [25]. These experiences revealed that radiosurgery could achieve long-term tumor growth control with very low morbidity, including facial palsy with hearing preservation. Recently, the combination of surgery and radiosurgery has become one of the effective treatment strategies for skull base meningiomas that allows preservation of cranial nerve functions while reducing morbidity [10,15,23]. We generally plan subtotal resections of the tumors and subsequent radiosurgery for large acoustic neuromas. The tumor volume is reduced by microsurgical removal, and the pre-existing neurologic deficits are improved, while long-term tumor growth control can be achieved by radiosurgery. Our experience is very small and we need more follow-up time. We could achieve 71% rate of no facial palsy after operation. In the final follow-up, we obtained 85.7% excellent facial nerve function (H&B Grade I or II). In our experiences, increased surgical experiences markedly reduced the risk of facial palsy. It is very important to avoid the extensive manipulation of the ventral site of the tumor capsule. Our tumor growth control was 79% and, among the patients with tumor growth, 1 patient had stabilized tumor size 5 years after radiosurgery. Only 1 patient (7%) with a very large hypervascular tumor underwent an operation after radiosurgery, but in this patient the tumor was too large and was not appropriate for radiosurgery. Interestingly, the tumor vascularity was substantially diminished at the last operation. Prasad et al reported 89% tumor growth control with previously operated patients and 94% tumor growth control with nontreated patients [25]. The tumor growth control rate with previously operated patients was worse compared to the nonoperative patients. We have 1 patient with subtotal removal without radiosurgery who we have followed for 3 years. But we think the residual tumor may be best treated by
Iwai et al
radiosurgery because of the low morbidity, less invasiveness, and high tumor growth control rate. In tumors with cyst formation, we recommend early radiosurgery after surgical resection. In the cases of large acoustic neuromas, subtotal removal and subsequent radiosurgery is a useful treatment strategy for maintaining cranial nerve function and long-term tumor growth control. REFERENCES 1. El-Kashlan HK, Zeitoun H, Arts HA, Hoff JT, Telian SA. Recurrence of acoustic neuroma after incomplete resection. Am J Otol 2000;21:389 –92. 2. Flickinger JC, Konziolka D, Niranjan A, Lunsford DL. Results of acoustic neuroma radiosurgery: an analysis of 5 years’ experience using current methods. J Neurosurg 2001;94:1–6. 3. Gardner G, Robertson JH. Hearing preservation in unilateral acoustic neuroma surgery. Ann Otol Rhinol Laryngol 1988;97:55–66. 4. Givre A, Olivecrona H. Surgical experience with acoustic tumors. J Neurosurg 1949;6:396 –407. 5. Harner SG, Ebersold MJ. Management of acoustic neuromas, 1978-1983. J Neurosurg 1985;63:175–9. 6. Horrax G. A comparison of results after intracapsular enucleation and total expiration of acoustic tumors. J Neurol Neurosurg Psychiatry 1950;13:268 –70. 7. House WF. Acoustic neuroma. Case summaries. Arch Otolaryngol 1968;88:586 –91. 8. House WF, Hitselberger WE. The neuro-otologist‘s view of the surgical management of acoustic neuromas. Clin Neurosurg 1985;32:214 –22. 9. House JW, Brackmann DE. Facial nerve grading system. Otolaryngol Head Neck Surg 1985;93:146 –7. 10. Iwai Y, Yamanaka K, Nakajima H. The treatment of skull base meningiomas— combining surgery and radiosurgery. J Clin Neurosci 2001;8:528 –33. 11. Iwai Y, Yamanaka K, Nakajima H. Two-staged gamma knife radiosurgery for the treatment of large petroclival and cavernous sinus meningiomas. Surg Neurol 2001;56:308 –14. 12. Jung S, Kang SS, Kim TS, et al. Current surgical results of retrosigmoid approach in extra large vestibular schwannomas. Surg Neurol 2000;53:370 –8. 13. Kameyama S, Tanaka R, Honda Y, Hasegawa A, Yamazaki H, Kawaguchi T. The long-term rate of residual acoustic neurinomas. Acta Neurochir (Wien) 1994;129:127–30. 14. Konziolka D, Lunsford LD, Mclaughlin MR, Flickinger JC. Long-term outcomes after radiosurgery for acoustic neuromas. N Engl J Med 1998;339:1426 –33. 15. Kurita H, Sasaki T, Kawamoto S, et al. Role of radiosurgery in the management of cavrenous sinus meningiomas. Acta Neurol Scand 1997;96:297–304. 16. Lanman TH, Brackmann DE, Hitselberger WE, et al. Report of 190 consecutive cases of large acoustic tumors (vestibular schwannoma) removed via the translabyrinthine approach. J Neurosurg 1999;90: 617–23. 17. Leksell L. A note on the treatment of acoustic tumours. Acta Chir Scand 1971;137:763–5. 18. Lownie SP, Drake CG. Radical intracapsular removal
Treatment for Large Acoustic Neuromas
19. 20. 21.
22.
23.
24.
25. 26.
27.
28.
of acoustic neurinomas. Long-term follow-up review of 11 patients. J Neurosurg 1991;74:422–5. Malis L. Gamma surgery for vestibular schwannoma. J Neurosurg 2000;92:894 –5 (letter). McKenzie KG, Alexander E. Acoustic neuroma. Clin Neurosurg 1955;52:21–36. Ohta S, Yokoyama T, Nishizawa S, Uemura K. Regrowth of residual tumours after acoustic neurinoma surgery. Br J Neurosurg 1998;12:419 –22. Olivecrona H. Analysis of results of complete and partial removal of acoustic neuromas. J Neurol Neurosurg Psychiatry 1950;13:271–2. Pendl G, Schrottner O, Eustacchio S, et al. Cavernous sinus meningiomas—what is the strategy: upfront or adjuvant gamma knife surgery? Stereotac Func Neurosurg 1998;70(Suppl 1):33–40. Pendl G, Unger F, Papaefthymiou G, Eustacchio S. Staged radiosurgical treatment for large benign cerebral lesions. J Neurosurg (Suppl 3) 2000;93:107–12. Prasad D, Steiner M, Steiner L. Gamma surgery for vestibular schwannoma. J Neurosurg 2000;92:745–59. Samii M, Turel KE, Penkert G. Management of seventh and eighth nerve involvement by cerebellopontine angle tumors. Clin Neurosurg 1985;32:242–72. Samii M, Matthies C. Management of 1000 vestibular schwannomas (acoustic neuromas): surgical management and results with an emphasis on complications and how to avoid them. Neurosurgery 1997;40:11–20. Yasargil MG, Smith RD, Gasser JC. Microsurgical approach to acoustic neurinomas. In: Krayenbu ¨ hl H, ed. Advances and Technical Standards in Neurosurgery, Vol. 4. Wien: Springer-Verlag, 1977:93–129.
COMMENTARY
Iwai et al present a retrospective study on a small series of 14 patients with large vestibular schwannomas. To reduce the risk of cranial nerve morbidity, they employed a combination of subtotal tumor resection followed by postoperative gamma knife radiosurgery an average of 3 months later. They achieved 71% early and 85% late (1–3 years) normal or near-normal facial nerve function (HouseBrackmann Grade 1 or 2) after treatment. One patient required reoperation; otherwise the residual tumor remained stable for up to 3 years. The early overall results in this series were favorable, but the question remains whether or not this concept is really acceptable or should even be recommended in certain patients. There are two points worthy of discussion: (1) subtotal resection and (2) the effect of radiosurgery on tumor control and morbidity. The idea of performing subtotal resection for the sake of obtaining a good functional outcome is not new and has been debated particularly with regard to facial nerve function, mainly before the era of neurophysiological monitoring. There was a period
Surg Neurol 289 2003;59:283–91
in which neuromonitoring was determined to be useful and in which complete resection with preservation of facial nerve function were the generally accepted goals. This development was supported by the experience that functional cochlear nerve function was also feasible. Along with these refinements and the frequently unsolvable problem of impending bilateral deafness in cases of neurofibromatosis Type 2, the idea of preserving function with subtotal resection was reintroduced and has proven successful [4]. However, it must be stressed that those indications are very rare and are reserved for patients with a real risk of bilateral deafness. Nonetheless, there has been a revival of subtotal resection for facial nerve preservation, and analysis has shown an inverse correlation between the extent of tumor resection and facial nerve preservation [3]. There is no doubt, however, that subtotal surgery will lead to regrowth and unacceptable outcome for a benign disease with a high rate of cure with complete resection. High rates of tumor recurrence with complete or partial tumor resection (9.2% vs. 20%) alone have been described [5], and while early recurrence was seen within 1 year, recurrence has also been seen unexpectedly after 6 or 8 years. Are there any clear criteria indicating the need for a subtotal resection, other than the individual surgeon’s ability and experience? To our knowledge, there are very few. The clinical facial function and the reliable neurophysiological evaluation are the most important factors. In cases where there is seen to be cancerous change in the facial nerve intraoperatively—as is commonly seen in NF2—and normal or good facial nerve function, the tumorous nerve will usually be left in continuity and function. As cure is not possible for NF2 patients, there is some doubt about whether radiosurgery in those cases is necessary or useful. As long as normal EMG responses can be elicited during tumor resection, normal facial nerve function, or at least some function with good chance of recovery, can be anticipated, and the surgeon may confidently continue the tumor resection. The authors do not mention any such individual considerations in their series. There have been trials combining surgery with radiosurgery [2] in which the regimen was determined to be effective in small groups of patients. Two points must be considered when prescribing radiosurgery after subtotal resection: the follow-up period and the double burden of morbidity. As mentioned above, recurrences or regrowth, particularly
Surgery Combined With Radiosurgery of Large Acoustic Neuromas Yoshiyasu Iwai, M.D., Kazuhiro Yamanaka, M.D., and Tomoya Ishiguro, M.D. Department of Neurosurgery, Osaka City General Hospital, Osaka, Japan
Iwai Y, Yamanaka K, Ishiguro T. Surgery combined with radiosurgery of large acoustic neuromas. Surg Neurol 2003;59:283–91. BACKGROUND
The treatment of acoustic neuromas has been improved by advancements in microsurgical techniques and in radiosurgery. To further elucidate the degree of clinical improvement, we evaluated the treatment results of a combination of surgery and radiosurgery for large acoustic neuromas. METHODS
From January 1994 through December 2000, we treated 14 patients with large acoustic neuromas using a combination of surgery and radiosurgery. Of these, 8 were male and 6 were female patients, with an average age of 47 years (range, 18 – 64). The average maximum diameter of the tumor was 42 mm (range, 30 –58 mm). All patients underwent operations using the retrosigmoid approach, and one patient was retreated using the transpetrosal transtentorial approach. The tumors were removed subtotally in thirteen patients and partially in one who had a very large hypervascular acoustic neuroma. There were no mortality and no surgical complications, such as hemorrhage or CSF leakage. Postoperative facial palsy was avoided in 10 patients (71%). Radiosurgery was performed 1 to 6 months (mean, 2.9 months) after surgery. At the time of radiosurgery, the treatment size (mean diameter) became 19.2 mm (range, 9.8 –36.1 mm). The average tumor marginal dose was 12.1Gy (range, from 10 –14 Gy). The mean follow-up period was 32 months after radiosurgery. RESULTS
The tumor size decreased in 6 patients, unchanged in 5 patients, and increased in 3 patients. Only 1 patient (7%) with extra large tumor needed surgical resection 1 year after radiosurgery. Excellent facial nerve function (House & Brackmann Grade I or II) was preserved in 12 patients (85.7%) in the final follow-up. CONCLUSIONS
In the case of large acoustic neuromas, subtotal removal and subsequent radiosurgery is one option for maintaining cranial nerve function and long-term tumor growth control. © 2003 Elsevier Inc. All rights reserved. Address reprint requests to: Dr Yoshiyasu Iwai, Department of Neurosurgery, Osaka City General Hospital, 2-13-22, Miyakojima-hondohri, Miyakojima-ku, Osaka, 534-0021, Japan. Received April 26, 2002; accepted November 20, 2002. © 2003 Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010 –1710
KEY WORDS
Vestibular schwannoma, surgery, radiosurgery, acoustic neuroma.
or the treatment of acoustic neuromas, excellent surgical results have been reported recently [27]. However, in terms of functional outcome, the surgical results for large acoustic neuromas may be worse compared to small or medium-sized tumors. Therefore, for large acoustic neuromas, intracapsular removal is one of the techniques used by some surgeons to preserve cranial nerve function [18]. However, the cases of subtotal removal have revealed a high recurrence rate [1]. We innovated the idea of intracapsular removal, thus allowing subtotal removal combined with radiosurgery, to increase the odds of preserving cranial nerve function and maintaining long-term tumor growth control.
F
Materials and Methods In our institute since January 1994, with a gamma knife unit (Elekta Instruments, Atlanta, Georgia, AL), we have taken maximum advantage of radiosurgery in the treatment of acoustic neuromas. We treated our series of acoustic neuromas in the following manner: The small- to medium-sized tumors with extrameatal diameters less than 3 cm were treated by radiosurgery, while the large tumors (defined as lesions with maximum extrameatal diameters greater than 3 cm) with neurologic symptoms, such as cerebellar ataxia and trigeminal neuropathy and other symptoms besides hearing disturbances, were treated by a combination of surgery and radiosurgery. Because these large tumors tend to cause clinical symptoms in addition to hearing disturbances, they must be resected to improve this symptomatology. 0090-3019/03/$–see front matter doi:10.1016/S0090-3019(03)00025-9
284 Surg Neurol 2003;59:283–91
Iwai et al
Preoperative Cranial Nerve Deficits and Clinical Signs Noted in 14 patients with Large Acoustic Neuromas
1
CLINICAL SIGNS 5th nerve 8th nerve Cerebellar sign Papilledema
NO.
%
7 14 8 4
50 100 57 29
According to this treatment strategy, we operated on 17 patients and 14 of these patients underwent radiosurgery from January 1994 through December 2000. In that period, 3 surgery patients were followed without radiosurgery. Of these, 2 patients received radiosurgery thereafter, but these 2 patients were not included in this study because only short follow-ups were possible after radiosurgery. One patient was observed after surgical removal without subsequent radiosurgery in accordance with the patient’s request. In this patient, the residual tumor was unchanged during 3 years after operation. Thus, we evaluated 14 patients who had undergone combined treatment (operation and radiosurgery). The study population was composed of 8 women and 6 men. The age range was 18 to 64 years (mean: 47 years). Two patients suffered from NF2. All patients suffered complications other than hearing disturbances, such as fifth cranial nerve palsy in 7 patients (50%); cerebellar ataxia in 8 patients (57%); and papilledema in 4 patients (29%) (Table 1). The hearing disturbance was evaluated according to the Gardner and Robertson (G&R) classification’s system [3]. There was one Class 1 patient, two Class 2 patients, two Class 3 patients, five Class 4 patients, and four Class 5 patients. Useful hearing (G&R class 1 or 2) was preserved in only 3 patients (21%). The maximum tumor diameter at the extrameatal portion was 30 to 40 mm in 6 patients, 40 to 50 mm in 6 patients, and above 50 mm in 2 patients (Table 2). The operation was performed via a retrosigmoid approach in all patients. In 1 patient with a very large hypervascular tumor associated with NF2, the
2
The Size Distribution of 14 Patients with Large Acoustic Neuromas
TUMOR SIZE (MM) 30–40 40–50 50⬍
NO.
%
6 6 2
43 43 14
tumor could not be removed by this approach because of cerebellar edema; thus, the transpetrosal transtentorial approach was employed. The operation was performed in the supine position. The tumor was removed in the following manner: The CSF was released from the cerebellopontine angle cistern. The arachnoid covering the tumor was incised and dissected. The dorsal part of the tumor was stimulated by a nerve stimulator to ensure that the facial nerve was not running in or near the dorsal portion of the tumor. The internal decompression of the tumor was performed by using the ultrasound aspiration and suction. The resection of the tumor capsule was performed from the cerebellar surface to the junction of the cerebellum and brainstem. The tumor around the internal auditory meatus and anterior portion close to the brain stem was left in place because manipulation of this part of the tumor could injure the facial nerve. The portion of tumor close to the lower cranial nerve was left in cases where the tumor dissection from these nerves was difficult. After tumor removal, the dura was sutured tightly and the suture site was sealed by fibrin-soaked Gelfoam to prevent postoperative CSF leakage, and the bone was replaced and fixed by a titanium plate. The interval between surgery and radiosurgery was 1 to 6 months (mean: 2.9 months). While the tumor seemed to adhere closely to the brainstem on image immediately after operation, after 3 months the tumor became a more appropriate shape for radiosurgery (Figure 1). In 2 patients with a cystic tumor, radiosurgery was performed 1 month after operation to prevent the growth of the cystic part of the tumor. In 1 patient with a very large hypervascular tumor (mean diameter: 58 mm), the radiosurgery was performed in a twostaged procedure with a 3-month interval between the stages, as we use for large skull base meningiomas [11,24]. The radiosurgery was performed for treatment size having diameters of 9.8 to 36.1 mm (mean: 18.9 mm) with tumor marginal doses ranging from 10 to 14.1Gy (mean: 12.1Gy). The patients could return to normal activity immediately after radiosurgery.
Results The tumor was removed subtotally in thirteen patients (93%) and partially in one patient (7%). The patient with partial removal had a very large hypervascular tumor, and we could not remove the tumor radically because of intraoperative bleeding. We did not experience operative mortality or other
Treatment for Large Acoustic Neuromas
Surg Neurol 285 2003;59:283–91
Tumor change after operation. A: Preoperative magnetic resonance imaging with gadolinium-enhancement showing a large acoustic neuroma at the left cerebellopontine angle in a 27-year-old woman. B: One month after operation, the residual tumor was situated at the left cerebellopontine angle. C: Three months after operation, the tumor shape became suitable for radiosurgery.
1
complications including CSF leakage or hemorrhage. One patient suffered trochlear nerve palsy. Facial palsy occurred in 4 patients (29%). The cause of the facial palsy cases was considered to be the
extensive manipulation of the ventral site of the tumor capsule, and the facial palsies were complete immediately after the operations. However, in the final follow-up, the facial palsy was improved to
286 Surg Neurol 2003;59:283–91
House and Brackmann (H&B) Grade II in 2 patients and Grade III in 2 patients [9]. We did not suffer other neurologic deficits after operations. Preexisting trigeminal neuropathy, cerebellar ataxia, and increasing intracranial pressure resolved in all patients during the early stages after the operations. Among 3 patients with useful hearing (with G&R Class 1 and 2 preoperatively), postoperatively useful hearing was preserved in only 1 patient. We were able to follow all patients in this study for 12 to 72 months (mean: 32 months) after radiosurgery. The tumors decreased in size in 6 patients (43%), remained unchanged in 5 patients (36%), and increased in size in 3 patients (21%) (Figure 2). Only 1 patient (7%) deteriorated because of tumor growth. In 1 patient with tumor growth, the tumor increased in size gradually during the 5 years after radiosurgery, and now remains stable in size. In 1 patient with NF2, the tumor was slightly increased in size during 24 months after radiosurgery. In the other patient (who had a very large hypervascular tumor with NF2), we could not remove the tumor radically, and this tumor with a mean diameter of 58 mm was treated by a two-staged radiosurgery procedure. In this patient, the tumor had been increasing in size and causing deterioration in neurologic condition; it was removed 1 year after radiosurgery. Interestingly, the tumor became avascular on angiographic imaging and was able to be removed subtotally. No other patient deteriorated after radiosurgery. No patient suffered radiation injury including facial palsy.
Discussion Lownie et al reviewed the historical surgical results of acoustic neuromas [18]. From their article, Cushing suffered significant mortality when attempting complete removal of the tumor, and he advocated intracapsular removal of the tumor. However, it was later demonstrated that such partial removals have a high recurrence rate [6]. Olivecrona et al reported good results of complete removal of the tumor in some patients and high rate of tumor regrowth after incomplete removal. But they still had high mortality for these procedures [4,22]. Mckenzie et al reported excellent results of intracapsular removal with low incidence of lower cranial nerve dysfunction, brain stem injury, and facial palsy [20]. Thus, microsurgical resection has been the primary treatment of acoustic neuromas since the microscope’s first use for acoustic neuroma surgery in 1961 [7]. Microsurgical resections allow radical tumor resections with less morbidity and mortality,
Iwai et al
but for large tumors the procedure still resulted in a high rate of facial palsy (37– 80%) in the 1980s [5,8,26,28]. Recently, excellent treatment results of surgery for acoustic neuromas have been reported. Samii reported 1,000 cases of surgery for acoustic neuromas and that his group could achieve total removals in 98% of cases. In addition, he noted facial nerve preservation in 87% of the patients [26]. Lamnan et al treated 190 patients who had large acoustic neuromas among 900 patients operated during the same period, 1990 through 1995, and achieved a total removal percentage of 96%. Low morbidity and excellent facial nerve function (H&B Grade I or II) was preserved (52.6%) in the patients on the final follow-up [16]. In the patients with extra large tumors (maximum extrameatal diameter above 4 cm), Jung et al reported their excellent surgical results with 73.3% of total removal and 86.4% of facial nerve preservation, and excellent facial nerve function (H & B Grade I or II) was preserved in 42.1% of patients [12]. So, the preservation of facial nerve function for the large acoustic neuromas was still difficult with radical removal of the tumors. Malis also commented that a surgeon needed the 200 cases to achieve excellent results with acoustic neuromas [19]. Lownie et al reported intracapsular removal of large acoustic neuromas in 11 patients with long follow-up for functional preservation during the era of microsurgery. The facial nerve was preserved in 9 patients (82%), but in 2 patients (18%) the residual tumor increased in size 2 and 3 years after surgery [18]. Kameyama et al followed 19 patients with residual tumor after intracapsular removal and 10 patients (53%) had regrowth. Among them, 5 patients with cyst formation had rapid regrowth (median, 4.5 years) and required reoperation, but the solid tumors were growing slowly (median, 15 years) [13]. Ohta et al reported that the tumor leaving the site of internal auditory canal had increased in size, and the tumor exiting the brainstem (and before entrance of internal auditory canal) had not increased during the follow-up period of 4.5 to 8 years [21]. El-Kashlan et al followed 39 patients with residual tumor for a mean period of 6.2 years, and the residual tumor was increased in size in 17 patients (44%) [1]. So, to reduce operative morbidity and preserve the functional outcome, intracapsular removal may be one operative approach in the era of microsurgery. But from the standpoint of tumor growth control, some adjuvant therapy may be needed for long-term tumor growth control. Radiosurgery has a long history of treating acous-
Treatment for Large Acoustic Neuromas
Surg Neurol 287 2003;59:283–91
A: Axial view of magnetic resonance (MR) imaging with gadolinium-enhancement showing a large tumor at the left cerebello-pontine angle compressing the brain stem of a 54-year-old female with clinical symptoms of left-sided deafness and cerebellar ataxia. B: The residual tumor after subtotal removal via the retrosigmoid approach without any neurologic deterioration. C: The radiosurgical treatment with a dose to the periphery of the tumor of 12 Gy. D: Four years after radiosurgery, MR imaging showing the tumor not having increased in size and the patient in good health with no new neurologic deficits.
2
tic neuromas. The first acoustic neuroma patient was treated in 1969 at Kalorinshka Hospital, using a gamma knife unit [17]. The treatment results were
improved by reducing the radiation dose and obtaining higher-quality images [2,14]. Konziolka et al reported the 5-year follow-up results in 165 patients
288 Surg Neurol 2003;59:283–91
and achieved 98% clinical tumor growth control (without tumor resection) [14]. Flickinger et al reported from the same institute as Konziolka that their tumor growth control rate was 91.0 ⫾ 2.5%. Few patients suffered facial palsy (1.1 ⫾ 0.8%) and 0% of those who received less that 13 Gy for tumor margin suffered facial palsy. Serviceable hearing was preserved in 73.5 ⫾ 4.7% of patients after 5 years [2]. Prasad and Steiner treated 159 patients with acoustic neuromas via gamma knife radiosurgery; they reported 89% tumor growth control in patients that had undergone previous operations and 94% tumor growth control in the nontreated patients along with 40% hearing preservation. Facial palsy occurred in only 2% and new trigeminal neuropathy occurred in 3% [25]. These experiences revealed that radiosurgery could achieve long-term tumor growth control with very low morbidity, including facial palsy with hearing preservation. Recently, the combination of surgery and radiosurgery has become one of the effective treatment strategies for skull base meningiomas that allows preservation of cranial nerve functions while reducing morbidity [10,15,23]. We generally plan subtotal resections of the tumors and subsequent radiosurgery for large acoustic neuromas. The tumor volume is reduced by microsurgical removal, and the pre-existing neurologic deficits are improved, while long-term tumor growth control can be achieved by radiosurgery. Our experience is very small and we need more follow-up time. We could achieve 71% rate of no facial palsy after operation. In the final follow-up, we obtained 85.7% excellent facial nerve function (H&B Grade I or II). In our experiences, increased surgical experiences markedly reduced the risk of facial palsy. It is very important to avoid the extensive manipulation of the ventral site of the tumor capsule. Our tumor growth control was 79% and, among the patients with tumor growth, 1 patient had stabilized tumor size 5 years after radiosurgery. Only 1 patient (7%) with a very large hypervascular tumor underwent an operation after radiosurgery, but in this patient the tumor was too large and was not appropriate for radiosurgery. Interestingly, the tumor vascularity was substantially diminished at the last operation. Prasad et al reported 89% tumor growth control with previously operated patients and 94% tumor growth control with nontreated patients [25]. The tumor growth control rate with previously operated patients was worse compared to the nonoperative patients. We have 1 patient with subtotal removal without radiosurgery who we have followed for 3 years. But we think the residual tumor may be best treated by
Iwai et al
radiosurgery because of the low morbidity, less invasiveness, and high tumor growth control rate. In tumors with cyst formation, we recommend early radiosurgery after surgical resection. In the cases of large acoustic neuromas, subtotal removal and subsequent radiosurgery is a useful treatment strategy for maintaining cranial nerve function and long-term tumor growth control. REFERENCES 1. El-Kashlan HK, Zeitoun H, Arts HA, Hoff JT, Telian SA. Recurrence of acoustic neuroma after incomplete resection. Am J Otol 2000;21:389 –92. 2. Flickinger JC, Konziolka D, Niranjan A, Lunsford DL. Results of acoustic neuroma radiosurgery: an analysis of 5 years’ experience using current methods. J Neurosurg 2001;94:1–6. 3. Gardner G, Robertson JH. Hearing preservation in unilateral acoustic neuroma surgery. Ann Otol Rhinol Laryngol 1988;97:55–66. 4. Givre A, Olivecrona H. Surgical experience with acoustic tumors. J Neurosurg 1949;6:396 –407. 5. Harner SG, Ebersold MJ. Management of acoustic neuromas, 1978-1983. J Neurosurg 1985;63:175–9. 6. Horrax G. A comparison of results after intracapsular enucleation and total expiration of acoustic tumors. J Neurol Neurosurg Psychiatry 1950;13:268 –70. 7. House WF. Acoustic neuroma. Case summaries. Arch Otolaryngol 1968;88:586 –91. 8. House WF, Hitselberger WE. The neuro-otologist‘s view of the surgical management of acoustic neuromas. Clin Neurosurg 1985;32:214 –22. 9. House JW, Brackmann DE. Facial nerve grading system. Otolaryngol Head Neck Surg 1985;93:146 –7. 10. Iwai Y, Yamanaka K, Nakajima H. The treatment of skull base meningiomas— combining surgery and radiosurgery. J Clin Neurosci 2001;8:528 –33. 11. Iwai Y, Yamanaka K, Nakajima H. Two-staged gamma knife radiosurgery for the treatment of large petroclival and cavernous sinus meningiomas. Surg Neurol 2001;56:308 –14. 12. Jung S, Kang SS, Kim TS, et al. Current surgical results of retrosigmoid approach in extra large vestibular schwannomas. Surg Neurol 2000;53:370 –8. 13. Kameyama S, Tanaka R, Honda Y, Hasegawa A, Yamazaki H, Kawaguchi T. The long-term rate of residual acoustic neurinomas. Acta Neurochir (Wien) 1994;129:127–30. 14. Konziolka D, Lunsford LD, Mclaughlin MR, Flickinger JC. Long-term outcomes after radiosurgery for acoustic neuromas. N Engl J Med 1998;339:1426 –33. 15. Kurita H, Sasaki T, Kawamoto S, et al. Role of radiosurgery in the management of cavrenous sinus meningiomas. Acta Neurol Scand 1997;96:297–304. 16. Lanman TH, Brackmann DE, Hitselberger WE, et al. Report of 190 consecutive cases of large acoustic tumors (vestibular schwannoma) removed via the translabyrinthine approach. J Neurosurg 1999;90: 617–23. 17. Leksell L. A note on the treatment of acoustic tumours. Acta Chir Scand 1971;137:763–5. 18. Lownie SP, Drake CG. Radical intracapsular removal
Treatment for Large Acoustic Neuromas
19. 20. 21.
22.
23.
24.
25. 26.
27.
28.
of acoustic neurinomas. Long-term follow-up review of 11 patients. J Neurosurg 1991;74:422–5. Malis L. Gamma surgery for vestibular schwannoma. J Neurosurg 2000;92:894 –5 (letter). McKenzie KG, Alexander E. Acoustic neuroma. Clin Neurosurg 1955;52:21–36. Ohta S, Yokoyama T, Nishizawa S, Uemura K. Regrowth of residual tumours after acoustic neurinoma surgery. Br J Neurosurg 1998;12:419 –22. Olivecrona H. Analysis of results of complete and partial removal of acoustic neuromas. J Neurol Neurosurg Psychiatry 1950;13:271–2. Pendl G, Schrottner O, Eustacchio S, et al. Cavernous sinus meningiomas—what is the strategy: upfront or adjuvant gamma knife surgery? Stereotac Func Neurosurg 1998;70(Suppl 1):33–40. Pendl G, Unger F, Papaefthymiou G, Eustacchio S. Staged radiosurgical treatment for large benign cerebral lesions. J Neurosurg (Suppl 3) 2000;93:107–12. Prasad D, Steiner M, Steiner L. Gamma surgery for vestibular schwannoma. J Neurosurg 2000;92:745–59. Samii M, Turel KE, Penkert G. Management of seventh and eighth nerve involvement by cerebellopontine angle tumors. Clin Neurosurg 1985;32:242–72. Samii M, Matthies C. Management of 1000 vestibular schwannomas (acoustic neuromas): surgical management and results with an emphasis on complications and how to avoid them. Neurosurgery 1997;40:11–20. Yasargil MG, Smith RD, Gasser JC. Microsurgical approach to acoustic neurinomas. In: Krayenbu ¨ hl H, ed. Advances and Technical Standards in Neurosurgery, Vol. 4. Wien: Springer-Verlag, 1977:93–129.
COMMENTARY
Iwai et al present a retrospective study on a small series of 14 patients with large vestibular schwannomas. To reduce the risk of cranial nerve morbidity, they employed a combination of subtotal tumor resection followed by postoperative gamma knife radiosurgery an average of 3 months later. They achieved 71% early and 85% late (1–3 years) normal or near-normal facial nerve function (HouseBrackmann Grade 1 or 2) after treatment. One patient required reoperation; otherwise the residual tumor remained stable for up to 3 years. The early overall results in this series were favorable, but the question remains whether or not this concept is really acceptable or should even be recommended in certain patients. There are two points worthy of discussion: (1) subtotal resection and (2) the effect of radiosurgery on tumor control and morbidity. The idea of performing subtotal resection for the sake of obtaining a good functional outcome is not new and has been debated particularly with regard to facial nerve function, mainly before the era of neurophysiological monitoring. There was a period
Surg Neurol 289 2003;59:283–91
in which neuromonitoring was determined to be useful and in which complete resection with preservation of facial nerve function were the generally accepted goals. This development was supported by the experience that functional cochlear nerve function was also feasible. Along with these refinements and the frequently unsolvable problem of impending bilateral deafness in cases of neurofibromatosis Type 2, the idea of preserving function with subtotal resection was reintroduced and has proven successful [4]. However, it must be stressed that those indications are very rare and are reserved for patients with a real risk of bilateral deafness. Nonetheless, there has been a revival of subtotal resection for facial nerve preservation, and analysis has shown an inverse correlation between the extent of tumor resection and facial nerve preservation [3]. There is no doubt, however, that subtotal surgery will lead to regrowth and unacceptable outcome for a benign disease with a high rate of cure with complete resection. High rates of tumor recurrence with complete or partial tumor resection (9.2% vs. 20%) alone have been described [5], and while early recurrence was seen within 1 year, recurrence has also been seen unexpectedly after 6 or 8 years. Are there any clear criteria indicating the need for a subtotal resection, other than the individual surgeon’s ability and experience? To our knowledge, there are very few. The clinical facial function and the reliable neurophysiological evaluation are the most important factors. In cases where there is seen to be cancerous change in the facial nerve intraoperatively—as is commonly seen in NF2—and normal or good facial nerve function, the tumorous nerve will usually be left in continuity and function. As cure is not possible for NF2 patients, there is some doubt about whether radiosurgery in those cases is necessary or useful. As long as normal EMG responses can be elicited during tumor resection, normal facial nerve function, or at least some function with good chance of recovery, can be anticipated, and the surgeon may confidently continue the tumor resection. The authors do not mention any such individual considerations in their series. There have been trials combining surgery with radiosurgery [2] in which the regimen was determined to be effective in small groups of patients. Two points must be considered when prescribing radiosurgery after subtotal resection: the follow-up period and the double burden of morbidity. As mentioned above, recurrences or regrowth, particularly