Introduction
Symptoms of vestibular schwannoma vary from hearing loss, tinnitus, vertigo, headache,
gait ataxia, facial nerve paresis to lower cranial nerve paresis. While much focus
in the literature is currently on functional preservation of cranial nerves and Gamma
Knife treatment, visual impairment secondary to vestibular schwannoma had been hardly
described in the published series. Neurosurgeons in developing countries commonly
encounter large and giant tumors presenting with visual impairment.
Only 19.4% of the vestibular schwannomas in the United States are of large size (≥
3 cm); giant tumors (> 4 cm) are even rarer (2–12.5%).[1]
[2]
[3] Unlike the developed world, more than 50% of vestibular schwannomas present to the
developing world neurosurgeons are either large or giant.[4] The situation is even worse in remote regions. In India, the state of Chhattisgarh
has one neurosurgeon per 3,887 km2 area and 8 lac population. Most neurosurgeons and tertiary care facilities are clustered
in the capital city. This distribution pattern makes access to tertiary healthcare
facilities difficult for patients living in remote regions who seek neurosurgical
help only when their symptoms have crippled them.
Visual impairment in vestibular schwannoma is generally attributed to raised intracranial
pressure due to hydrocephalus. However, there are a subset of patients without hydrocephalus
who develop visual impairment. Other hypotheses for visual impairment are also described
in the literature. We intend to present our experience and describe the likely causes
of visual impairment in patients with vestibular schwannomas in developing countries,
like India. This study also aims to highlight this overlooked entity of the visual
impairment secondary to vestibular schwannoma.
Materials and Methods
We conducted a retrospective review of our data from January 2015 to December 2020
(6 years). There were 156 patients operated on for vestibular schwannoma. A total
of 42 patients were excluded (13 patients do not have preoperative visual records,
and 29 patients have refractive errors). The final analysis included 114 patients.
The clinical and radiological data were retrieved from medical record department.
The follow-up of the patients having preoperative visual impairment was conducted
in August to September 2021, and informed consent was obtained.
A total of eight parameters were studied—age, gender, maximum tumor size in cm, presence
or absence of hydrocephalus, economic status, duration of symptoms, visual acuity
status, and ease of access to the tertiary neurosurgical facility.
The tumor was categorized into (A) tumor size less than 4 cm and (B) 4 cm or larger
in maximum dimension. Hydrocephalus was defined by Evan's index more than 0.3. Economic
status was categorized into below poverty line and above poverty line (APL). Visual
impairment/loss was defined by World Health Organization (WHO) criteria (< 6/18).
Visual status was categorized into four groups—(A) normal vision (6/6–6/18), (B) visual
impairment (< 6/18–6/60), (C) severe visual impairment (< 6/60–3/60), and (D) blind
(< 3/60—no perception of light) on Snellen's chart. Visual impairment attributed to
the tumor was considered when associated with concomitant fundus findings and onset
of visual impairment after primary symptom of vestibular schwannoma. Access to the
tertiary neurosurgical facility was dichotomized based on geographical location into
easy—capital city Raipur and adjoining districts (7 districts) and difficult—remote
districts (21 districts).
Statistical analysis was done using JASP software (version 0.14.1, Amsterdam). Continuous
and categorical variables were expressed as mean ± standard deviation. For categorical
variables, chi-square and Fisher's exact test were used. Continuous variables were
analyzed by the independent t-test. Stepwise logistic regression was applied to observe the influence of variables
on outcome. Two-tailed p-values were considered statistically significant at p-value less than 0.05.
Results
The final analysis included 114 patients, of which 42 (36.84%) patients had a varying
degree of visual impairment (visual acuity < 6/18). Positive fundus findings were
present in 79/114 (69.29%) cases (papilloedema in 71 and secondary optic atrophy in
8 patients). Visual impairment (WHO category B) was seen in 14.91%, severe visual
impairment (WHO category C) was present in 12.28% of cases, while 9.64% of patients
were blind (WHO category D) ([Table 1]). For the statistical analysis, visual functions were dichotomized into impaired
vision (WHO category B, C, D) and normal vision (WHO category A).
Table 1
Preoperative visual status of the patients with postoperative outcome
|
Preoperative status (n = 114)
|
Postoperative improvement (n = 14/35)[a]
|
|
Normal vision (6/6–6/18)
|
72 (63.15%)
|
–
|
|
Visual impairment (< 6/18–6/60)
|
17 (14.91%)
|
10/15 (66.67%)
|
|
Severe visual impairment (< 6/60–3/60)
|
14 (12.28%)
|
03/11 (27.27%)
|
|
Blind (< 3/60–no perception of light)
|
11 (9.64%)
|
01/09 (11.11%)
|
a 7 patients—3 dead, 4 lost to follow-up.
The mean age of the total cohort was 48.76 ± 9.97 (range: 22–69) years. There was
no significant difference in the groups (impaired vision vs. normal vision) with respect
to age, gender, and economic status ([Table 2]).
Table 2
Statistical analysis of variables for visual impairment in patients having vestibular
schwannoma
|
Factors
|
Overall
|
Impaired vision
|
Normal vision
|
p-Value
|
|
Age (y)
|
48.76 ± 9.97 (22–69)
|
47.83 ± 9.84 (22–67)
|
49.30 ± 10.08 (26–69)
|
0.45 (NS)
|
|
Gender (M/F)
|
52 (45.61%): 62 (54.38%)
|
25 (59.52%): 17 (40.47%)
|
37 (51.38%): 35 (48.61%)
|
0.40 (NS)
|
|
DOS (mo)
|
17.44 ± 6.16 (6–36)
|
19.14 ± 5.99 (9–36)
|
16.45 ± 6.08 (6–30)
|
0.02 (S)
|
|
Maximum tumor size (cm)
|
4.19 ± 0.44 (3.3–5.1)
|
4.31 ± 0.46 (3.4–5.1)
|
4.12 ± 0.41 (3.3–5.0)
|
0.02 (S)
|
|
Tumor size category (A: < 4 cm/B: ≥ 4 cm)
|
40 (35.08%): 74 (64.91%)
|
12 (28.57%): 30 (71.42%)
|
28 (38.88%): 44 (61.11%)
|
0.26 (NS)
|
|
Economic status (BPL/APL)
|
89 (78.07%): 25 (21.93%)
|
35 (83.33%): 07 (16.66%)
|
54 (75%): 18 (25%)
|
0.30 (NS)
|
|
Hydrocephalus (yes/no)
|
64 (56.14%): 50 (43.86%)
|
29 (69.04%): 13 (30.95%)
|
35 (48.61%): 37 (51.38%)
|
0.03 (S)
|
|
Ease of access (easy/difficult)
|
42 (36.84%): 72 (63.15%)
|
9 (21.42%): 33 (78.57%)
|
33 (45.83%): 39 (54.16%)
|
0.009 (S)
|
Abbreviations: APL, above poverty line; BPL, below poverty line; DOS, duration of
symptoms; F, female; M, male.
The mean duration of symptoms was significantly higher in patients with impaired vision
(19.14 vs. 16.45 months, p = 0.02). Although the patients with impaired vision had significantly larger tumors
(4.31 vs. 4.12 cm, p = 0.02), when tumors are categorized into two groups (< 4 cm vs. ≥ 4 cm), no significant
difference was observed (p = 0.26) ([Table 2]).
Hydrocephalus was present in 64/114 (56.14%) cases, and patients with hydrocephalus
had higher chances of visual impairment (p = 0.03). Overall, there were 42/114 (36.84%) patients from Raipur and adjoining districts,
and the rest, 72/114 (63.15%), patients were from remote districts. Visual impairment
was significantly higher in patients from remote districts (p = 0.009) ([Table 2]). A total of 28/64 (43.75%) patients underwent preoperative ventriculoperitoneal
(VP) shunt placement, 22 in the impaired vision group and 6 in normal vision group.
Visual impairment was significantly associated with four factors—larger tumor size,
longer duration of symptoms, presence of hydrocephalus, and difficult access to the
neurosurgical facility. These factors were further analyzed in stepwise logistic regression
analysis, longer duration of symptoms, and poor access to tertiary neurosurgical facility
stand out as decisive factors for visual impairment. Larger tumor size and presence
of hydrocephalus failed to sustain their significance ([Table 3]).
Table 3
Stepwise logistic regression analysis of studied variable: dependent variable—visual
loss (present/absent), covariates—age, tumor size, DOS, factors—gender, tumor size
category, economic status (BPL/APL), HCP (present/absent), ease of access (easy/difficult)
|
Factors
|
Odds ratio (OR)
|
Z-score
|
p-Value
|
|
Age
|
0.98
|
−0.65
|
0.51 (NS)
|
|
Gender (male)
|
0.72
|
−0.74
|
0.45 (NS)
|
|
Duration of symptoms
|
1.07
|
2.15
|
0.03 (S)
|
|
Maximum tumor size
|
1.45
|
0.37
|
0.70 (NS)
|
|
Tumor size category (> 4 cm)
|
0.55
|
−1.02
|
0.30 (NS)
|
|
Economic status (BPL)
|
1.19
|
0.32
|
0.74 (NS)
|
|
Hydrocephalus (present)
|
1.52
|
0.88
|
0.37 (NS)
|
|
Ease of access (difficult)
|
3.41
|
2.62
|
0.009 (S)
|
Abbreviations: APL, above poverty line; BPL, below poverty line; DOS, duration of
symptoms; HCP, hydrocephalus; NS, nonsignificant; S, significant.
Intercept included; null model applied.
The follow-up data were available for 35/42 (83.33%) cases of impaired vision group,
three patients were dead, and four were lost to follow-up. Visual improvement was
defined as an upgrade to better categories. Overall, there was a visual improvement
after surgery in 14/35 (40%) of cases (66.67% in WHO category B, 27.27% in WHO category
C, and 11.11% in WHO category D) ([Table 1]). Vision improved in a significantly higher number of patients when a preoperative
VP shunt was placed (p = 0.03) ([Table 4]).
Table 4
Cross-tabulation of postoperative visual outcomes of the patient with respect to the
VP shunt placement
|
Vision improved
|
Vision not improved
|
p-Value
|
|
VP shunt placed
|
12
|
10
|
0.03 (S) Fisher's exact test
|
|
VP shunt not placed
|
2
|
11
|
Abbreviation: VP, ventriculoperitoneal.
Discussion
Problem Statement
The incidence of visual impairment secondary to vestibular schwannoma is challenging
to determine. Among the western series, the incidence ranges from 1 to 10%, while
in developing world series, it varies from 6.2 to 57.5% ([Table 5]).[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10] Actual incidences may differ due to underreporting. In the Western world, patients
seek medical attention early at the onset of symptoms; therefore, small tumors are
detected early.[11] In a large European series of 1,865 patients, the mean tumor size was 2.2 cm in
the late 1980s,[12] while in a contemporary Indian series of 510 patients over 43 years, most of the
tumors were large as patients here seek medical help only when sufficiently disabled.[13] This Indian series was categorized into three eras. Most of the patients were blind
at presentation in the first decade, and 16% were blind in the middle era. Although,
after the availability of the computed tomography (CT) scan (CT era), smaller tumors
were picked up earlier, more than 50% had papilledema.[13] This report highlights the availability of neurosurgical facilities and the importance
of imaging that is still a distant dream in rural areas, which comprises nearly two-third
of the Indian population.[14] In our series, more than 50% of patients had papilloedema at presentation. It seems
not much has changed in the rural and tribal India of the twenty-first century; we
still get nearly one-third (36.84%) of such cases with visual impairment.
Table 5
Review of the literature of developing and developed world describing visual impairment
in patients of vestibular schwannoma
|
Author
|
n
|
Tumor size range (mm)
|
Mean tumor size (mm)
|
Hydrocephalus
|
Papilloedema
|
Visual impairment
|
Visual improvement
|
|
Studies from the developing world
|
|
Jain et al 2005[4]
|
259
|
11–> 40
|
–
|
–
|
116 (44.78%)
|
36 (13.89%)
|
–
|
|
Kumar et al 2013[5]
|
40
|
> 30
|
–
|
22 (55%)
|
32 (80%)
|
23 (57.5%)
|
–
|
|
Huang et al 2013[10]
|
1,009
|
T3–T4
|
38
|
–
|
–
|
82 (8.1%)
|
–
|
|
Nair et al 2016[6]
|
64
|
27–66
|
41
|
40 (62.5%)
|
20 (31.2%)
|
04 (6.2%)
|
–
|
|
Turel et al 2016[7]
|
179
|
40–67
|
–
|
155 (86.59%)
|
128 (71.50%)
|
48 (26.81%)
|
–
|
|
Present study
|
114
|
33–51
|
41.9
|
64 (56.14%)
|
79 (69.29%)
|
42 (36.84%)
|
14/35 (40%)
|
|
Studies from the developed world
|
|
van Meter et al 1983[8]
|
100
|
5–65
|
–
|
–
|
8 (8%)
|
02 (2%)
|
–
|
|
Matthies and Samii 1997[9]
|
1,000
|
11–45
|
33
|
–
|
–
|
1–3%
|
–
|
|
Samii et al 2010[3]
|
50
|
40–65
|
44
|
16 (43.2%)
|
13 (26%)
|
05 (10%)
|
–
|
The referral pattern and presentation to the tertiary neurosurgical facility are an
area of concern. It is unfortunate but true that patients with hearing loss in rural
areas either self-medicate or get treated by nonregistered medical practitioners.
Similarly, tinnitus and vertigo do not get adequately evaluated. Visual impairment
often gets attributed to anterior chamber disease, mostly cataracts in elderly patients.
These patients do not have easy access to ophthalmologists and otologists with adequate
diagnostic tools in the vicinity. As long as the patient has one normal ear and visual
impairment that does not significantly disable day-to-day activities, symptoms are
ignored, the tumor keeps growing, and intracranial pressure keeps rising. These arguments
are in concordance with the observations highlighted in one Indian study.[15] Developing world's neurosurgeons encounter and operate mostly large or giant tumors,
with the outcomes destined poor, compared with the western literature. The results
of our study also highlight the significantly higher incidence of visual loss in patients
from remote areas and with a longer duration of symptoms.
According to the estimates of the financial express, nearly 22% of the Indian population
is below the poverty line. This is even worse in Chhattisgarh, where it is 45%.[16] In our study, we could not find an association between visual impairment and economic
status. The likely reason is that our institute is the only state government-run tertiary
care neurosurgical center that mostly provides treatment covered by government schemes;
thus, practically all patients belong to the poor economic group.
It is apparent that the neurosurgeons of the developed world might not have significant
cases with visual impairment due to smaller tumors, early presentation, early detection,
and ease of access to a proper facility. The situation in developing world, on the
contrary, is otherwise, and the neurosurgeons treat a higher proportion of large and
giant tumors. Larger tumors have higher chances of having concomitant hydrocephalus
and thereby a higher number of visually handicapped patients. Reporting of visual
impairment gets outshined by the priority given to facial and hearing preservation
efforts. We believe almost all neurosurgeons of developing world must have encountered
patients of vestibular schwannomas with varying degrees of visual loss. Here, we emphasize
to essentially include complete ophthalmological assessment in the preoperative workup
and its evaluation postoperatively.
Reasons for Visual Loss
Tumor Size
We evaluated the tumor size in relation to visual loss. The mean tumor size was significantly
larger in patients having visual loss (4.31 vs. 4.12 cm). When the tumors were grouped
into large and giant size categories, the risk of visual loss was not different. This
finding is in concordance with other studies. Huang et al in their study of 1,009
vestibular schwannomas reported no difference in visual loss between T3 and T4 categories.[10] Similarly, no difference was observed between the tumor size less than 5 cm and
more than or equal to 5 cm by Turel et al.[7] van Meter et al observed that in patients with visual findings, tumor size was at
least 4 to 4.5 cm.[8] Intracranial pressure-related symptoms have been found significantly higher in tumor
size more than 4 cm.[17] From these reports, one may conclude that visual impairment sets in after a certain
tumor size (likely 4 cm) and further growth does not change the visual status statistically.
Hydrocephalus/Increased Intracranial Pressure
Large and giant tumors compress the fourth ventricle leading to obstructive hydrocephalus.
Hydrocephalus is seen even with smaller tumors suggesting communicating etiology.[18] In a series of 167 patients of Sammi et al, 2.4% of patients had hydrocephalus with
a mean tumor size of 2.3 cm.[3] Visual impairment and papilloedema may occur even without ventriculomegaly. Matos
et al reported a case of optic disc edema without hydrocephalus in a 3 cm tumor.[19] Papilledema without hydrocephalus is also described in an NF 2 case.[20] Similar observations were reported in other studies.[21] A giant tumor may present with papilloedema and visual impairment without hydrocephalus.[22] These reports suggest that hydrocephalus is not mandatory for the development of
papilloedema in vestibular schwannoma. As long as any cerebrospinal fluid (CSF) cleft
persists inside the fourth ventricle, CSF circulation remains intact.
Our univariate analysis, our results suggest that larger tumor size, hydrocephalus,
longer duration of symptoms, and poor ease of access to tertiary facility are the
factors responsible for visual loss. However, on a multivariate logistic regression
analysis, longer duration of symptoms and poor ease of access were main decisive factors.
These findings reflect that larger tumor size and hydrocephalus are resultant of delayed
presentation, and by improving ease of access, we can prevent visual impairment.
CSF Hyperproteinorrachia
Kumar et al demonstrated significantly higher protein values from the CSF of cisterna
magna in vestibular schwannoma patients, which correlated significantly with visual
impairment. In logistic regression analysis, duration of symptoms, papilloedema, hydrocephalus,
and tumor volume was not significant for visual loss.[5] Shedding of protein through the tumor capsule was the probable cause. Protein concentration
in ventricular CSF was also higher in their study. Bloch et al postulated that chronic
inflammation, abnormal blood-brain barrier, or repeated subarachnoid hemorrhage from
tumor surface vessels might lead to an increase in fibrinogen content of CSF, which
ultimately gets converted into fibrin at the level of arachnoid granulations.[23] There is another possibility that vestibular schwannoma may produce local cisternal
and ventricular stasis of CSF flow, which leads to an increase in the protein levels,
as seen in Froin's syndrome. Higher protein contents in CSF may also produce coagulum,
which may occlude CSF flow across arachnoid granulations, resulting in an early increase
in intracranial pressure without early radiological hydrocephalus.[23]
[24]
[25] Fukuda et al, in their study, found high CSF protein as a main responsible factor
for hydrocephalus, and as the tumor grows, CSF protein level rises.[26] This cause and effect relationship is proven by the work of Miyakoshi et al, which
demonstrated normalization of CSF protein contents and pressure after tumor removal.[18]
Gray Matter Alterations
Long-term hearing impairment may produce gray matter structural alterations even outside
the auditory cortex. Wang et al, in their volumetric evaluation by MRI of 42 patients
with unilateral hearing loss due to vestibular schwannoma, found decreased gray matter
volume in the calcarine cortex.[27] Functional connectivity in the visual cortex gets altered in patients with unilateral
hearing loss.[28]
[29] These findings suggest that visual impairment may occur even in the absence of raised
intracranial pressure.
Prognosis
We could not find any literature on the prognosis of visual impairment after the primary
tumor has been dealt with. In our study, 40% of patients had improvement in vision.
Preoperative VP shunt placement helped improve vision. Patients with visual impairment
need thorough ophthalmological assessments to rule out correctable causes of blindness,
like anterior chamber disease. Once there is evidence of increased intracranial pressure
on imaging, either in the form of hydrocephalus or tortuous and dilated intraorbital
perioptic CSF spaces, early surgery is indicated. In the circumstances of heavy caseload
like in a government hospital, at least CSF diversion procedures should be done while
awaiting definitive surgery.[7] We preoperatively performed a CSF diversion procedure in 28/64 (43.75%) of patients
with hydrocephalus. Jefferis et al have successfully demonstrated the benefit of optic
nerve sheath fenestration in three patients of visual impairment secondary to vestibular
schwannoma.[30]
Quality of Life
There are numerous studies on quality of life in patients of vestibular schwannoma.[31]
[32]
[33]
[34]
[35]
[36]
[37] The quality-of-life assessment consists of an evaluation of the patient's physical
health, psychological state, and level of independence, social relationships, and
relationships to their own environment. None of the published studies included visual
impairment as a parameter for quality-of-life assessment.[38] The visually handicapped patients remain dissatisfied and live miserable lives.
Once the primary devil, “the tumor,” has been dealt with, the focus of these patients
shifts to functional independence, and they often leave neurosurgeons baffled with
the questions such as: “Why did you operate to give me a blind life?”[13] The surgery in these unfortunate patients does save lives, but the survivors live
with the agony of deafness, imbalance, and facial weakness; blindness adds salt to
their wounds.
Recommendations
It is evident from the present study that socioeconomic and geographical factors do
play an essential role in delayed presentation. Schuz et al have suggested improving
community awareness about early symptoms, specifically in the less educated and less
affluent society.[39] We need to strengthen the people's belief in the neurosurgical services offered
by the government hospitals, which, unfortunately, lag behind in self-promotion. The
neurosurgeons need to break out of their shells to reach out to the people in remote
areas either through continuing medical education (CMEs) or camps and need to gain
the trust of their otology and ophthalmology colleagues, whom these patients consult
first. Due to the lower incidence of vestibular schwannomas and poor availability
of imaging facilities in remote areas, routine screening with imaging is impractical.
Rupa et al have suggested evaluation with auditory brain stem response over magnetic
resonance imaging as a cost-effective screening method in patients having asymmetric
audio-vestibular symptoms in the hospital settings.[40] Early diagnosis of vestibular schwannoma is the preventive measure for visual impairment.
Managing these large tumors with limited facilities requires courage and skills. It
is the need of the hour for neurosurgical associations to work hand in hand with the
government and create opportunities by establishing proper neurosurgical infrastructure
in rural areas so that freshly graduated neurosurgeons may be motivated to provide
“rural neuroscience” services.
Strength and Limitations of the Study
The possible reasons for visual loss in posterior fossa lesions are well documented
in the literature. Our results validate these observations. To the best of our knowledge,
no published study had solely focused on visual impairment in vestibular schwannomas
and its prognosis after treatment. Our study shows the chances of recovery in visual
impairment after treatment is 40%, which could help in patient prognostication. This
study highlights the existing lacunae in the healthcare delivery system in the developing
countries. We believe the results of our study will raise awareness in the scientific
community.
The main limitation of the study is its retrospective nature. We could not get the
pretreatment fundus examination photographs to compare with the present status. In
many patients, imaging was done at other center, which prevented us doing the exact
volumetric analysis of tumors. For this reason, we took maximum tumor size as a parameter.
In the literature, both tumor size and tumor volume had been used to assess the neurological
outcomes.
Conclusions
There are a significant number of visually handicapped patients having vestibular
schwannoma. The primary reason is delayed arrival at the tertiary neurosurgical facility
due to poor ease of access in developing countries. We need to strengthen our healthcare
system for early detection. The scenario of vestibular schwannomas in developing world
is vastly different from that of the developed world, and patient's eyesight may still
be salvaged if timely evaluated and addressed.
