This article summarizes available evidence on various management options for vestibular schwannoma as they relate to the decision-making strategies used in selection. After a brief consideration of individual options, the literature directly comparing two or more management options is examined, noting the level of evidence supporting their claims. A discussion of the strategies developed to guide decision making follows. The article closes with a summary of the evidence-based findings and suggestions for further research. The focus is on management of sporadic, unilateral vestibular schwannomas, because patients with neurofibromatosis type 2 pose different management problems best discussed separately.
| CPA | Cerebellopontine angle |
| GBI | Glasgow Benefit Inventory |
| QOL | Quality of life |
| RCTs | Randomized controlled trials |
| SRS | Stereotactic radiosurgery |
| SRT | Stereotactic fractionated radiation therapy |
| VS | Vestibular schwannoma |
Decision-making strategies for management of vestibular schwannomas (VS) reflect this tumor’s benign, slow-growing natural history (see the article by Stangerup elsewhere in this issue). Treatment strategies typically emphasize maintenance of quality of life (QOL) and preservation of neurologic function. Radical tumor resection at the expense of neighboring neural tissue is only undertaken when the benefits of surgery, on some realistic time scale, outweigh the anticipated cost in QOL. These tenets have become increasingly important as smaller tumors are discovered earlier and with fewer preoperative neurologic deficits.
There are three main management strategies for VS: (1) observation, (2) radiation treatment, and (3) surgery. Given the natural history of VS, observation with periodic examination and imaging is an acceptable option for many patients and in developed countries has become the dominant initial management strategy for most smaller tumors. When active intervention is indicated, alternatives are microsurgery, which includes retrosigmoid (see the article by Morcos and Telischi elsewhere in this issue), translabyrinthine (see the article by Arriaga and Lin elsewhere in this issue), and middle fossa (see the article by Angeli elsewhere in this issue) approaches, and targeted radiation therapy (see the article by Link and colleagues elsewhere in this issue), including stereotactic radiosurgery (SRS) and stereotactic fractionated radiation therapy (SRT).
Arguments favoring one management strategy over the others hinge on a variety of outcome measures, related to the aforementioned goal of lesion management with preservation of neurologic function and QOL. One important end point or measure of “success” is tumor control or stability, although the precise definition of this end point differs among the three strategies. In the case of observation, it suggests absence of measureable growth or growth slow enough to warrant continued withholding of active intervention; for microsurgery, it implies completeness of resection without recurrence, or failure of residual, unresected tumor to grow after surgery; and for targeted radiation therapy, it denotes freedom from future growth of irradiated tumor. The most common neurologic outcome measures are preservation of hearing and facial function, given the tumor’s origin from cranial nerve (CN) VIII and its proximity to CN VII. Balance and tinnitus (CN VIII) and facial sensation (CN V) are other occasionally tested functions. Recent studies have also focused on QOL per se, as measured by various subjective questionnaire instruments, and financial cost to society.
In this article, the reader finds the following:
- •
Available evidence on the various management options as they relate to the decision-making strategies used to select between them
- •
Consideration of individual management options
- •
Literature review directly comparing two or more management options, noting the level of evidence supporting their claims
- •
Discussion of the strategies developed to guide decision making
- •
Summary of the evidence-based findings and suggestions for further research.
The focus is on management of sporadic, unilateral VS, because patients with neurofibromatosis type 2 pose different management problems best discussed separately.
Observation
One strategy that has been widely adopted especially in the management of small VS in older patients is surveillance with regular clinical review and diagnostic imaging to evaluate for tumor growth. Traditionally, among the reasons for a trial of conservative management is advanced age (usually >65 years), where the extent of tumor growth in the patient’s lifetime is unlikely to mandate intervention. In patients whose medical history poses an increased risk of postoperative complications, a trial of conservative management may also be warranted. Small tumor size and those causing few, if any symptoms, especially when entirely intracanalicular, are often managed with observation. Further indications for observation are risk of deterioration of good hearing, or if the tumor is involving the better or only hearing ear. Some patients choose observation for smaller tumors with good hearing and switch preference to active management only if hearing is lost during observation. Patient preference is another reason that a patient may undergo a “wait and scan” option even for medium-size tumors.
Since 1985 there have been several studies using conservative management of VS. A summary of some of the larger recent studies is provided in Table 1 . In these, tumor growth ranged between 30% and 40% and regression between 2% and 19.4%. The mean annual growth rates of all tumors varied from 0.3 to 1.42 mm/yr. In studies that specifically mentioned growth rates of tumors needing treatment because of failure of conservative management, rates ranged from 3 to 4.8 mm/yr (see Table 1 ).
| Study | N | Follow-Up a (mo) | Growth Rate (mm/yr) | Failure Rate (%) | |
|---|---|---|---|---|---|
| Overall | Treated b | ||||
| Suryanarayanan et al, 2010 | 327 | 43 (12–168) | 1.1 | 4.8 | 24.5 |
| Bakkouri et al, 2009 | 386 | NA (12–108) | 1.15 | NA | 23.7 |
| Godefroy et al, 2009 | 70 | 40 (11–73) | 0.45 | NA | 39 |
| Malhotra et al, 2009 | 202 | 30 (1–192) | NA | NA | 9.4 |
| Ferri et al, 2008 | 123 | 58 (6–182) | 0.3 | NA | 36.3 |
| Hajioff et al, 2008 | 72 | 121 (80–271) | 1 | NA | 35 |
| Stangerup et al, 2008 | 636 | 47 (4–137) | NA | NA | 20.8 |
| Raut et al, 2004 | 72 | 80 (52–234) | 1 | 3.1 | 32 |
| Hoistad et al, 2001 | 102 | 29 (6–120) | NA | NA | 33.3 |
| Walsh et al, 2000 | 72 | 40 (12–194) | 1.42 | 4.2 | 15.3 |
a Mean or median follow-up period, with range indicated in parentheses.
b Growth rate of tumors requiring intervention (either microsurgery or targeted radiation therapy).
Failure of conservative management was seen in 9.4% to 40% of patients in the various recently published studies with the mean follow-up ranging from 2.5 to 4.8 years. One study from Canada with the longest follow-up to date (median, 10 years) has shown a failure rate of 35%, of which 75% failed within the first 5 years of follow-up. Reasons for failure of conservative management include tumor growth rate greater than or equal to 3 mm/yr; increasing cochleovestibular symptoms (disabling vertigo, hearing deterioration); and patient choice.
As seen from the studies, almost two-thirds of patients show favorable results with the conservative “wait and scan” management option. For many patients the conservative approach is a reasonable initial management of VS, and in the authors’ clinic it is the most frequent initial choice for patients with smaller tumors.
Microsurgery for VS
The aim of microsurgery of VS is to accomplish complete tumor removal while preserving CN function and brainstem and cerebellar integrity. Since the introduction of microsurgical VS resection in the early 1960s, studies have demonstrated tumor control rates of approximately 95%, irrespective of approach. A summary of several of the larger surgical series is provided in Table 2 .
| Study | N | Large a (%) | Follow-Up b (mo) | Approach | GTR (%) | RR (%) | CN VII c (%) | CN VIII d (%) | |
|---|---|---|---|---|---|---|---|---|---|
| Anat. | Func. | ||||||||
| Lanman et al, 1999 | 190 | 100 | 12 | TL | 96.3 | NA | 93.7 | 52.6 | 0 |
| Samii & Matthies, 1997 | 1000 | 80 | NA | RS | 97.9 | 0.8 | 93 | NA | 39.5 |
| Gormley et al, 1997 | 179 | 59 | 65 (3–171) | RS | 99 | 1 | 98 | NA | 38 |
| Sampath et al, 1997 | 611 | NA | NA | TL/RS | 99.5 | 0.82 | 97.5 | 89.7 | NA |
| Buchman et al, 1996 | 96 | NA | ≥12 | RS | 77 | 6 | 95 | 75 | 48 |
| Glasscock et al, 1993 | 161 | 46 | 78 (12–144) | RS/MF | 100 | 1 | 99 | NA | 36 |
| Ebersold et al, 1992 | 255 | 62 | NA | RS | 97 | 3 | 92.6 | 64 | 19 |
a Fraction of patients with tumors >3 cm.
b Mean or median follow-up period, with range indicated in parentheses.
c Rate of facial nerve preservation.
Recurrence rate after microsurgery is highly dependent on the extent of removal. In the largest series to date, Samii and Matthies have shown that 979 of 1000 tumors were completely resected, with a recurrence rate of 0.8%. Total resection rates of 97% to 99% are associated with recurrence rates of 0% to 3%.
Recent microsurgical studies have shown that in more than 90% of patients there is complete tumor removal, with anatomic and functional preservation of the facial nerve. The risk of developing facial nerve damage from microsurgery depends on the tumor size. Gormley and colleagues found that the likelihood of a favorable outcome (House Brackmann grade I–II) was 96% for tumors smaller than 2 cm 3 , 74% for tumors 2 to 4 cm 3 , and 38% for tumors larger than 4 cm 3 . Other studies have also supported this relationship of greater preservation rates with smaller tumor size.
Hearing preservation after microsurgery has presented a greater challenge compared with preserving the facial nerve. Preservation of hearing is critically dependent on the size of the tumor. Functional hearing preservation (Gardner-Robertson Class I–II) was seen in 48% of tumors smaller than 2 cm 3 , 25% in tumors 2 to 4 cm 3 , and none in those larger than 4 cm 3 . Other surgical series have reported a 14% to 50% rate of preservation of serviceable hearing (pure tone average <50 dB, speech discrimination score >50%). Recurrence of tumor after complete resection is rare in most series.
Microsurgery for VS
The aim of microsurgery of VS is to accomplish complete tumor removal while preserving CN function and brainstem and cerebellar integrity. Since the introduction of microsurgical VS resection in the early 1960s, studies have demonstrated tumor control rates of approximately 95%, irrespective of approach. A summary of several of the larger surgical series is provided in Table 2 .
| Study | N | Large a (%) | Follow-Up b (mo) | Approach | GTR (%) | RR (%) | CN VII c (%) | CN VIII d (%) | |
|---|---|---|---|---|---|---|---|---|---|
| Anat. | Func. | ||||||||
| Lanman et al, 1999 | 190 | 100 | 12 | TL | 96.3 | NA | 93.7 | 52.6 | 0 |
| Samii & Matthies, 1997 | 1000 | 80 | NA | RS | 97.9 | 0.8 | 93 | NA | 39.5 |
| Gormley et al, 1997 | 179 | 59 | 65 (3–171) | RS | 99 | 1 | 98 | NA | 38 |
| Sampath et al, 1997 | 611 | NA | NA | TL/RS | 99.5 | 0.82 | 97.5 | 89.7 | NA |
| Buchman et al, 1996 | 96 | NA | ≥12 | RS | 77 | 6 | 95 | 75 | 48 |
| Glasscock et al, 1993 | 161 | 46 | 78 (12–144) | RS/MF | 100 | 1 | 99 | NA | 36 |
| Ebersold et al, 1992 | 255 | 62 | NA | RS | 97 | 3 | 92.6 | 64 | 19 |
a Fraction of patients with tumors >3 cm.
b Mean or median follow-up period, with range indicated in parentheses.
c Rate of facial nerve preservation.
Recurrence rate after microsurgery is highly dependent on the extent of removal. In the largest series to date, Samii and Matthies have shown that 979 of 1000 tumors were completely resected, with a recurrence rate of 0.8%. Total resection rates of 97% to 99% are associated with recurrence rates of 0% to 3%.
Recent microsurgical studies have shown that in more than 90% of patients there is complete tumor removal, with anatomic and functional preservation of the facial nerve. The risk of developing facial nerve damage from microsurgery depends on the tumor size. Gormley and colleagues found that the likelihood of a favorable outcome (House Brackmann grade I–II) was 96% for tumors smaller than 2 cm 3 , 74% for tumors 2 to 4 cm 3 , and 38% for tumors larger than 4 cm 3 . Other studies have also supported this relationship of greater preservation rates with smaller tumor size.
Hearing preservation after microsurgery has presented a greater challenge compared with preserving the facial nerve. Preservation of hearing is critically dependent on the size of the tumor. Functional hearing preservation (Gardner-Robertson Class I–II) was seen in 48% of tumors smaller than 2 cm 3 , 25% in tumors 2 to 4 cm 3 , and none in those larger than 4 cm 3 . Other surgical series have reported a 14% to 50% rate of preservation of serviceable hearing (pure tone average <50 dB, speech discrimination score >50%). Recurrence of tumor after complete resection is rare in most series.
Stereotactic radiosurgery
Since its introduction in 1969 as a treatment for VS, SRS has been applied increasingly more often, because of favorable demonstrated outcomes and because it offers an outpatient alternative to an inpatient procedure requiring general anesthesia. SRS offers single fraction dose administration and thus short treatment times, which is convenient for patients. LINAC radiation therapy (using a linear accelerator adapted for SRS) is an alternative to gamma knife radiosurgery and similar outcomes have been seen for both in terms of tumor control and side effects. A summary of several recent series is presented in Table 3 .
| Study | N | Marginal Dose (Gy) | Follow-Up a (mo) | Control b (%) | CN V c (%) | CN VII d (%) | CN VIII e (%) | Failure (%) | |
|---|---|---|---|---|---|---|---|---|---|
| Median | Range | ||||||||
| Lobato-Polo et al, 2009 | 55 | 13 | 11–20 | 64 (48–240) | 96 | 96.4 | 98.2 | 87 | 5.5 |
| Chopra et al, 2007 | 216 | 13 | 12–13 | 68 | 98.3 | 94.9 | 100 | NA | 1.4 |
| Hasegawa et al, 2005 | 317 | 13.2 | 10–18 | 94 | 92 | NA | 98 | 68 | 6.9 |
| Hasegawa et al, 2005 | 73 | 14.6 | 10–18 | 135 | 87 | NA | 89 | 68 | 12.3 |
| Flickinger et al, 2004 | 313 | 13 | 12–13 | 24 | 98.6 | 95.6 | 100 | 78.6 | 0.6 |
| Iwai et al, 2003 | 51 | 11.7 | 8–12 | 60 (18–96) | 92 | 100 | 100 | 56 | 4 |
| Spiegelmann et al, 2001 | 44 | 14.6 | 11–20 | 32 (12–60) | 98 | 82 | 76 | 71 | 2.3 |
| Petit et al, 2001 | 47 | 12 | 7.5–14 | 43 (12–84) | 96 | 100 | 96 | 88 | 4.2 |
| Foote et al, 2001 | 149 | 14 | 10–22.5 | 34 (6–94) | 93 | 71/98 f | 71/95 f | NA | 4 |
| Flickinger et al, 2001 | 190 | 13 | 11–18 | 30 | 91 | 97.4 | 98.9 | 74 | 5 |
| Prasad et al, 2000 | 153 | 13 | 11–18 | 52 (12–120) | 92 | 97 | 98 | 65 | NA |
a Mean or median follow-up period, with range indicated in parentheses.
b Rate of tumor growth control, often representing actuarial results.
c Rate of trigeminal nerve preservation.
d Rate of facial nerve preservation.
e Rate of hearing preservation.
f Different values refer to different time periods in the study (before/after 1994).
Indications for SRS have included advanced age, failure of prior microsurgery, patient preference, and medical comorbidities precluding general anesthesia. SRS has also been preferred for smaller tumors because of the increased risk of trigeminal and facial nerve neuropathies after radiation treatment of larger tumors, and increased risk of tumor swelling causing posterior fossa crowding and obstructive or communicating hydrocephalus. Studies have tended to use size criteria of less than 3 cm in mean diameter or less than 15 cm 3 in volume.
As opposed to microsurgery, in which the goal is tumor resection, the goal of SRS is tumor control, or freedom from progression of tumor growth. In the early Swedish experience, of 14 tumors ranging in size from 7 to 30 mm, eight (57%) decreased in size, two (14%) remained stable, and three (21%) increased. In a large series of 829 patients treated by the Pittsburgh group, long-term tumor control (absence of the need for further intervention) was 98% in the 252 patients with 10-year follow-up or longer. Of these, 73% experienced a decrease in tumor volume, and 25% had no change. Notably, however, these results reflect changing radiation doses during the first 5 years of their experience. From initial tumor margin doses of 18 to 20 Gy, adapted from the early Swedish experience, the dose gradually decreased to 14 to 16 Gy by 1992. Their results during this early period, in which the mean dose was 16.6 Gy, showed a high overall growth control rate of 98%, with a decrease in size in 62%, no change in 53%, and increase in 6%. Since 1992 they used a margin dose of 12 to 13 Gy, and reported actuarial 6-year control rates of 98.6% in 313 patients with median 24 months of follow-up.
Another study with comparable follow-up duration (over 10 years) to that by Kondziolka and colleagues, but using a lower mean margin dose of 14.6 Gy, showed a lower control rate of 87% overall, and 93% in tumors smaller than 10 cm 3 . A later study by the same group in 346 patients with median follow-up of 7.8 years and lower mean margin dose of 13.2 Gy showed actuarial 10-year control rates of 92% overall, and 97% in tumors smaller than 15 cm 3 . Treatment failures observed in these studies tended to occur within the first 3 years. A summary of recent SRS studies is shown in Table 3 , with reported failure rates (defined by tumor progression) ranging from 0.6% to 12%. Thus, whereas the early studies using higher marginal doses showed a high rate of tumor control, longer follow-up is needed to confirm whether such rates are durable using lower doses.
The risk of CN neuropathy is similarly margin dose-dependent. Foote and colleagues stratified 149 patients by marginal dose level and showed that those receiving 20 or 22.5 Gy experienced a 100% rate of CN (V or VII) neuropathy, whereas those receiving 10 or 12.5 Gy experienced a rate of 0% and 2.9%, respectively. Dichotomizing results with a cutoff of 12.5 Gy, those receiving less had a 2-year incidence of CN V or VII neuropathy of 2%, whereas those receiving more had an incidence of 24%.
Consistently, studies using higher doses have reported lower rates of facial nerve preservation. In one study of 162 patients receiving 16.6 Gy, facial nerve preservation was 79%. Other studies using mean margin doses of 14.6 and 19.4 Gy reported rates of 76% and 79%, respectively. In contrast, studies using lower doses (mean, 12–13) have reported facial preservation rates ranging from 96% to 100% (see Table 3 ).
Similarly, rates of trigeminal nerve preservation show that higher doses (mean, 14.6–19.4 Gy) are associated with lower preservation rates of 73% to 86%, whereas lower doses (mean, 12–13 Gy) are associated with higher preservation rates of 95% to 100%. Hearing preservation rates show more variability, but follow the same trend. Rates using 14.6 to 19.4 Gy have ranged from 50% to 71%, whereas rates using 12 to 13 Gy have ranged from 56% to 88%.
As experience has accumulated during the past few decades, marginal tumor dose prescriptions have been lowered to reduce CN neuropathies. With magnetic resonance imaging replacing computed tomography and improved treatment-planning software, a large number of isocenters can be used to attain more precise conformality and sharper dose fall-off. The more recent studies have thus demonstrated less neurologic morbidity, but the question is whether this is achieved at a cost to tumor growth control. A comparison between two contemporary studies at different centers with similar long-term follow-up (≥10 years) suggests lower control rates (87% vs 98%) with lower margin dose (14.6 vs 16.6 Gy). Consistently, Pollock and colleagues reviewed their experience in 293 patients with minimum 24-month follow-up, and found that those treated before 1997 with an average marginal dose of 16 Gy had a 7-year actuarial tumor control rate of 98%, but those treated after 1997 with an average marginal dose of 12 Gy had a 90% control rate. Further studies and longer follow-up are needed to address this important issue.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
