Keywords
vestibular schwannoma - advanced age - microsurgery - radiosurgery
Introduction
Vestibular schwannoma (VS) are benign skull base neoplasms arising from the eighth
cranial nerve. Patients with sporadic VS often present with any of the combination
of hearing loss, tinnitus, dizziness, and/or gait instability. Converging evidence
suggests a rising incidence affecting upward of 20 per 100,000 person-years in persons
aged 70 years or older, possibly driven by increasing neuroimaging study rates.[1] Management of VS must account for facial nerve function, hearing preservation, quality
of life, and factor in age at the time of presentation. As there is no clear gold
standard, management of VS must be approached on a case by case basis and ideally
performed by experienced surgeons in high volume centers. Current management paradigms
for VS often incorporate either microsurgical resection or stereotactic radiosurgery
(SRS), however, in older patients especially, observation with serial monitoring may
be a preferred strategy due to the tumor's relatively indolent nature.[2]
[3]
[4]
The best available data suggests that the treatment-naive VS have an average estimated
annual volumetric growth rate of 33.5% per year.[5] Based on the same volumetric analysis, only one-third of the tumors remain stable
in size. The approach to VS in patients of advanced age must balance risk/benefit
considerations of intervention, given actuarial life expectancy predictions versus
anticipated tumor growth rates. It is for this reason that many high-volume centers
elect to observe smaller tumors that are either asymptomatic or mildly symptomatic,
and tend to treat larger, symptomatic tumors preferably using SRS in this older age
demographic. Although SRS can achieve tumor control in 94% of patients at 10 years,
only 69% achieve tumor control if the VS demonstrates growth of >2.5 mm per year prior
to treatment.[6]
[7]
We hypothesize that, when warranted or preferable, microsurgical resection of VS in
older patients can be performed with acceptable tumor control and facial nerve outcomes.
Further, some lesions are not optimal for radiosurgical intervention due to size or
volumetric constraints and proximity to the brainstem or cochlea. There is a paucity
in the literature detailing surgical outcomes of elderly patients undergoing microsurgical
resection of VS. We report amongst the largest surgical series focusing on outcomes
of microsurgical resection of VS in patients of advanced age.
Materials and Methods
A retrospective analysis of a prospectively maintained, institutional review board
(IRB) approved database was performed to identify sporadic (non-NF2) VS patients aged
65 years or older undergoing microsurgical resection from 1988 to 2020. All patients
provided consent to be included in the database. A cohort of the same age demographic
undergoing radiosurgery as the primary means of intervention was constructed for comparison.
Patient demographics, presenting symptoms, prior intervention, tumor characteristics,
surgical records, complications, disposition information, and follow-up data were
recorded.
Patients presenting to our institution with VS are treated by a multidisciplinary
team composed of otolaryngology, neurosurgery, and radiation oncology. Patients presenting
with incidentally found tumors, or presenting with minor symptoms, are offered observation,
radiosurgery, or surgical resection depending on tumor and patient factors. Symptomatic
patients are typically offered microsurgical resection versus radiosurgery. VS with
diameter >30 mm and/or those with a large cystic component are usually not considered
for radiosurgery. Patients in this series with tumor dimensions amenable to radiosurgery
were offered GKRS or surgery, but opted for surgical resection following extensive
counseling. Patient reasoning for preference of surgery over radiosurgery was recorded
when available.
Extent of resection was defined by comparing operative records with postoperative
magnetic resonance imaging at 3 months follow-up. Gross total resection (GTR) was
defined by complete microscopic removal of tumor with no evidence of disease on MRI.
Near total resection (NTR) was defined if residual tumor no greater than 25 mm2 and 2 mm thick along the facial nerve or brainstem was present, as described by Bloch
et al.[8] All other forms of incomplete resection were deemed subtotal resections (STR). In
agreement with most large reported case series, we defined recurrence as tumor growth
of >2 mm on follow-up imaging.[9]
[10]
Facial nerve outcomes were defined using the House Brackmann (HB) scale and subcategorized
as “good” outcomes being HB1–2, “moderate” HB3–4, and “poor” HB5–6. HB scores were
obtained within the first week postoperatively prior to the time of discharge, and
again at the time of most recent follow-up. Rates of facial nerve improvement were
defined by improvement of subcategory from the time of discharge compared with the
most recent clinical follow-up. A linear regression was performed in RStudio using
the “glm” function to test for correlation between facial nerve outcomes and tumor
size.
Results
Demographics and Presentation
Sixty-four patients met inclusion criteria and were identified for analysis. There
were another 62 VS patients of the same age demographic opting for radiosurgery as
opposed to microsurgical resection. Average age at the time of surgery was 71.2 years,
ranging from 65 to 84 years. The cohort was composed of 62.5% women (n = 40) and 37.5% men (n = 24). Presenting symptoms included 84.4% hearing loss (n = 54), 48.4% vertigo (n = 31), 10.9% gait ataxia from brainstem compression (n = 7), 3.1% facial numbness (n = 2), and 1.6% trigeminal neuralgia (n = 1). These results are summarized in [Table 1].
Table 1
Patient demographics and presentation
|
Surgical cohort (n = 64)
|
|
Average age (years)
|
71.2 [65–84]
|
|
Sex
|
|
|
Male
|
37.5% (n = 24)
|
|
Female
|
62.5% (n = 40)
|
|
Presentation
|
|
|
Hearing loss
|
84.4% (n = 54)
|
|
Vertigo
|
48.4% (n = 31)
|
|
Gait imbalance
|
10.9% (n = 7)
|
|
Facial numbness
|
3.1% (n = 2)
|
|
Trigeminal neuralgia
|
1.6% (n = 1)
|
Surgical Outcomes
Microsurgical resection was the primary means of intervention in all patients. Patients
with tumor diameter exceeding 30 mm (42.4%, n = 27) were typically not considered for radiosurgery. Similarly, patients with a
large cystic component were rarely offered radiosurgery based on our institutional
experience and preference (28.1%, n = 18). Nearly three-quarters (n = 45) of patients in this series were offered radiosurgery or surgical resection,
and ultimately opted for surgery. The most common patient reasoning for electing surgery
included preference to not undergo radiation (n = 31, 48.4%). A small subset of patients desired a “surgical cure” in hopes of receiving
an expedited relief of symptoms (n = 6, 9.4%). One patient was not offered radiosurgery as he presented with acute symptom
onset due to intratumoral hemorrhage necessitating surgical resection.
Mean maximum tumor diameter was 29 mm, ranging from 13 to 55 mm. A translabyrinthine
approach was used in 73.4% (n = 47) of cases, with the remaining 26.6% (n = 17) performed via a retrosigmoid approach. GTR was achieved in 39.1% (n = 25) of cases, NTR in 32.8% (n = 21), and STR in 28.1% (n = 18). These results are summarized in [Table 2].
Table 2
Surgical data, tumor characteristics, and extent of resection
|
Surgical outcomes
|
|
Tumor size (maximum diameter, cm)
|
29 [13–55]
|
|
Cystic
|
28.1% (n = 18)
|
|
Indication for Surgery
|
|
|
Refused GKRS
|
48.4% (n = 31)
|
|
“Surgical cure”
|
9.4% (n = 6)
|
|
Large (>3cm)
|
42.2% (n = 27)
|
|
Intratumoral hemorrhage
|
1.6% (n = 1)
|
|
Approach
|
|
|
Translabyrinthine
|
73.4% (n = 47)
|
|
Retrosigmoid
|
26.6% (n = 17)
|
|
EOR
|
|
|
GTR
|
39.1% (n = 25)
|
|
NTR
|
32.8% (n = 21)
|
|
STR
|
28.1% (n = 18)
|
Abbreviations: EOR, extent of resection; GTR, gross total resection; NTR, near total
resection; STR, subtotal resection.
Clinical Course
The average hospital length of stay was 5 days, ranging from 2 to 17 days. Only one
patient remained hospitalized over 1 week due to an aspiration event requiring reintubation
and eventual tracheostomy. Ultimate disposition at the time of discharge was home
in 75% (n = 48), acute rehabilitation unit in 18.8% (n = 12), skilled nursing facility in 4.7% (n = 3), and one mortality (1.5%). Complications included: dural sinus injury in 3.1%
(n = 2), cerebellar infarct 3.1% (n = 2), wound infection in 3.1% (n = 2), cerebrospinal fluid leak in 3.1% (n = 2), respiratory distress in 1.5% (n = 1) requiring reintubation, and one postoperative hemorrhage resulting in emergent
evacuation. The patient developing postoperative hemorrhage occurred during the first
postoperative night and resulting in a devastating neurological injury with ultimate
withdrawal of care on postoperative day five. These results are summarized in [Table 3].
Table 3
Hospital course, ultimate disposition, and resultant complications
|
Clinical course
|
|
Length of stay (days)
|
5 [2–17]
|
|
Disposition
|
|
|
Home
|
75%(n = 48)
|
|
ARU
|
18.8%(n = 12)
|
|
SNF
|
4.7% (n = 3)
|
|
Morgue
|
1.5% (n = 1)
|
|
Complication
|
|
|
Sinus injury
|
3.1% (n = 2)
|
|
Cerebellar infarct
|
3.1% (n = 2)
|
|
Wound infection
|
3.1% (n = 2)
|
|
CSF leak
|
3.1% (n = 2)
|
|
Respiratory distress
|
1.6% (n = 1)
|
|
Hemorrhage
|
1.6% (n = 1)
|
|
Death
|
1.6% (n = 1)
|
Abbreviations: ARU, acute rehabilitation unit; CSF, cerebrospinal fluid leak; SNF,
skilled nursing facility.
Facial Nerve Outcomes and Tumor Control
Average postoperative time of follow-up was 37.8 months. Tumor control was achieved
in all but three case (95.3%). All patients with notable postoperative tumor growth
were referred for SRS. Two patients were referred for adjuvant CyberKnife and one
for GKRS due to residual tumor along the facial nerve and brainstem at 4 months after
resection. These results are summarized in [Table 4].
Table 4
Follow-up data, tumor control, adjuvant radiosurgery, and facial nerve outcomes
|
Follow-up data
|
|
|
Follow-up duration (months)
|
|
|
Mean
|
37.8 [1–264]
|
|
Tumor control
|
|
|
Yes
|
95.3% (n = 61)
|
|
No
|
4.7% (n = 3)
|
|
Adjuvant SRS
|
|
|
GKRS
|
3.1% (n = 2)
|
|
CyberKnife
|
3.1% (n = 2)
|
|
HB (post-op week 1)
|
|
|
Good [1–2]
|
43.8% (n = 28)
|
|
Moderate [3–4]
|
32.8% (n = 21)
|
|
Poor [5–6]
|
23.4% (n = 15)
|
|
HB (most recent)
|
|
|
Good [1–2]
|
51.6 (n = 33)
|
|
Moderate [3–4]
|
31.3% (n = 20)
|
|
Poor [5–6]
|
17.1% (n = 11)
|
|
Facial nerve improvement
|
10.9% (n = 7)
|
Abbreviations: GKRS, gamma knife radiosurgery; HB, House Brackmann; SRS, stereotactic
radiosurgery.
Initial postoperative HB scores were assigned within the first week after surgery
prior to discharge. Postoperative week 1 HB scores were good (HB1–2) in 43.8% (n = 28) of patients, moderate (HB3–4) in 32.8% (n = 21), and poor (HB5–6) in 23.4% (n = 15) of cases. At the time of follow-up, patients demonstrated good HB scores in
51.6% (n = 33), moderate HB in 31.3% (n = 20), and poor HB in 10.9% (n = 7). When comparing facial nerve function from the first postoperative week to the
time of follow-up, only 10.9% (n = 7) made any form of improvement. These results are summarized in [Table 4]. There was no correlation between tumor size and facial nerve outcomes ([Fig. 1]).
Fig. 1 Linear regression demonstrating no significant correlation between presenting tumor
size and facial nerve outcomes.
Radiosurgery
During the same time period, an additional 62 VS patients of the same age demographic
opted for Gamma Knife radiosurgery as the primary means of intervention. Average age
at time of radiosurgery was 73 years, ranging from 65 to 86 years. The cohort had
a female predominance at 59.7% (n = 37). Mean maximum tumor diameter was significantly smaller than the microsurgical
cohort at 18.9 mm, ranging from 5 to 37 mm (p <0.001). All patients were treated at a dose of 12 to 13Gy. The most common presenting
symptom was hearing loss in 80.6% (n = 50), followed by gait instability in 17.7% (n = 11), facial numbness in 4.8% (n = 3), or as an incidental finding in 4.8% (n = 3). Only 6.5% (n = 3) had a cystic component to the tumor at the time of treatment. Average postoperative
time of follow-up was 30.8 months. Tumor control was achieved in all but three case
(95.2%). These results are summarized in [Table 5].
Table 5
Demographics, tumor characteristics, dosing, presenting symptoms, and follow-up duration
of the radiosurgery cohort being treated from 1988 to 2020
|
Radiosurgery cohort (n = 62)
|
|
Average age (years)
|
73 [65–86]
|
|
Sex
|
|
|
Male
|
40.3% (n = 25)
|
|
Female
|
59.7% (n = 37)
|
|
Average tumor size (maximum diameter, cm)
|
18.9 [5–37]
|
|
Dose
|
12–13Gy
|
|
Presentation
|
|
|
Hearing loss
|
80.6% (n = 50)
|
|
Gait instability
|
17.7% (n = 11)
|
|
Facial numbness
|
4.8% (n = 3)
|
|
Incidental
|
4.8% (n = 3)
|
|
Cystic
|
6.5% (n = 3)
|
|
Follow-up duration (months)
|
|
|
Mean
|
30.8 [1–154]
|
|
Tumor control
|
|
|
Yes
|
95.2% (n = 59)
|
|
No
|
4.8% (n = 3)
|
Discussion
Our findings from a large series of surgical resection of VS from a high volume, tertiary
care center suggest that, despite age, most patients can undergo resection with low
rates of complications, standard lengths of hospitalization, and a majority able to
be discharged home. In the current age of radiosurgery, many institutions preferably
employ SRS for tumor control in elderly patients.[11] We reject the notion that patients of advanced age should be universally precluded
from consideration of surgical resection, and show that in selected cases favorable
outcomes may be achieved in this patient population. We advocate that symptomatic
patients that are otherwise reasonable surgical candidates should be offered both
radiosurgery and microsurgical resection. This highlights the importance for VS being
managed at high volume tertiary institutions, where an emphasis on safety, adjuvant
treatment, and multidisciplinary care can tailor treatment plans based on patient
preferences and risk considerations.
There have been previous VS surgical series dedicated to reviewing outcomes in older
patients, though the majority of prior work was published prior to the widespread
availability and use of SRS. Silverstein et al[12] argued that large, symptomatic VS are best treated by STR to decompress the brainstem
while minimizing damage to the facial nerve and reduce postoperative morbidity. Samii
et al[13] advised that elderly patients can still undergo GTR with excellent outcomes, and
that age was not correlated with outcomes. Van Abel et al,[14] Bowers et al,[15] and Jiang et al[16] agreed by concluding that although older patients had poorer health at the time
of surgery, there was no resultant increase in morbidity profile. Our results are
congruent with previous reports in that we do not find increasing age to pose an elevated
risk profile. The complication profile of our current series is comparable to our
previous VS surgical series comprised of patients of all ages.[17] One recent study by McCutcheon et al[18] draws contrast to other works by finding older age to be associated with increased
morbidity, including stroke or respiratory failure. Older studies do exist that describe
a correlation between increasing age and morbidity/mortality, but many such studies
are at or prior to the widespread use of radiosurgery. In such series, surgeons did
not have the luxury of foregoing aggressive resections with reliance on radiosurgery
for tumor control.[19] As such, it is not unexpected that older patients would have experienced worse outcomes
resulting from more extensive resections. We did not find weaning the ventilator or
airway considerations to be major difficulties in most of our patients, with the exception
of one patient requiring re-intubation in the early postoperative period.
All patients included in our series were symptomatic at the time of presentation.
Our institution will offer surgical resection or radiosurgery to symptomatic patients
who are surgical candidates in the event that tumor maximum diameter does not exceed
3 cm. Outside of the patients included in this study, another 62 VS patients of the
same age demographic presenting to our institution opted for radiosurgery as opposed
to microsurgical resection. Thus, of the 126 patients presenting with symptomatic
VS over the age of 65 offered both radiosurgery or surgical resection, patients were
equally as likely to pick either option. The primary reason for the surgical cohort
to opt for resection was due to an aversion to radiosurgery or the desire to have
more rapid relief of symptoms, particularly gait imbalance, offered by surgical debulking.
On the other hand, the majority of patients preferring radiosurgery presented with
hearing loss only or had smaller tumors, often purely intracanalicular in nature.
The average size of tumor at the time of presentation was at the threshold for consideration
for radiosurgery while the patients opting for radiosurgery tended to have smaller
tumors, or purely intracanalicular tumors. Our institutional philosophy favors resection
in tumors >2.5 cm in maximum diameter, especially as a surgical approach offers an
expedited relief of symptoms caused by larger tumors. While there have been studies
supporting effective tumor control of VS >3 cm in a single dimension, or >10 cm3, tumor volume exceeding 15 cm3 is a significant predictive factor for poor tumor control following radiosurgery,
in addition to adverse radiation effects to the brainstem.[20] Interestingly, a quarter of our patients in this age demographic presented with
a cystic component. These patients would undergo a surgery mostly to relieve acute
symptom onset due to rapid expansion of the cyst causing mass effect on the brainstem.
There has been a great deal of interest in the quality of life as it pertains to VS
patients in recent years. The diagnosis of VS, as opposed to treatment modality chosen,
seems to have the biggest impact on the quality of life.[21]
[22] Symptoms most influential on lowering quality of life were ongoing headaches and
dizziness with facial nerve function and hearing loss less siginficant.[23] Patients experience improved quality of life following GTR as compared with having
residual tumor.[24] The culmination of these studies underscore the importance of appropriate patient
counseling and advocates for intervening on symptomatic patients in which an intervention
can be reasonably expected to relieve symptoms. This notion is perhaps most evident
in the elderly population, as the quality of life improvement from the relief of symptoms
outweighs to the burden of undergoing surgical resection.
Any discussion of VS treatment modalities and resultant outcomes must consider tumor
control and facial nerve preservation. The approach to an older patient with VS will
have somewhat different goals of surgery compared with a younger patient with longer
life expectancy. With respect to tumor control, the goals of surgery for large tumors
should be to decompress the brainstem to relieve symptoms and preserve neurological
function. Our institution achieved 64.3% rates of GTR across all VS cases performed
across the time span on this study; however, in the older patients we reserve more
aggressive resection to have rates of GTR or NTR as 39.1 and 32.8%, respectively.
Less aggressive attempted resections in our advanced age cohort perhaps also, at least
partially, explain why our linear regression did not find correlation between tumor
size and facial nerve outcomes. There have been multiple multivariate analyses failing
to show a correlation between age at the time of surgery as an independent risk factor
for poor facial nerve recovery.[25]
[26]
[27] In our series, very few patients had a meaningful improvement in their HB score
from the time of discharge to follow-up. Considering facial nerve function is a determinant
of postoperative quality of life, significant effort should be placed on maintaining
the integrity of the nerve.[28]
Limitations
This study is limited by its small sample size. However, it represents one of the
largest single center surgical series of VS patients over the age of 65 undergoing
microsurgical resection. Further, the follow-up duration of three years is short when
discussing VS. Some patients have less than a 1 year follow-up as the resections were
performed within the year of this study. However, the primary narrative of our findings
is that surgery can be performed in the advanced age demographic with good facial
nerve outcomes and a comparable short-term clinical course compared with younger patients.
We have no reason to suspect that long-term tumor control following STR would behave
differently due to patient's age. Lastly, we have omitted hearing outcomes from our
analysis. The query of institutional databases demonstrated incomplete records or
inconsistent methods of describing hearing status making meaningful analysis difficult.
Conclusion
We report a single institution's surgical series suggesting VS patients of advanced
age can still undergo microsurgical resection with very good outcomes. Surgical resection
in elder patients has the option of opting for less aggressive resections to decompress
the brainstem and preserve facial nerve function, as most tumors will not exhibit
growth in the patients' remaining lifetime. We reject the notion that elderly patients
with VS should not be considered for surgical candidacy and thus receive radiosurgery
as the primary means of intervention.
