Keywords
brain tumor - trigeminal schwannoma - microsurgery - radiosurgery
Introduction
Trigeminal schwannomas (TS) are benign neoplasms originating from the nerve sheath,
distal to the schwann–oligodendroglia junction.[1] These tumors of the skull base account for 0.07 to 0.36% of intracranial tumors
and 1 to 8% of adult intracranial schwannomas, but they can make up around 10% of
schwannomas encountered in adolescence.[2]
[3]
[4]
[5]
[6] This correlates to an approximate incidence of 1 case per 100,000 inhabitants, affecting
predominantly women between the fourth and sixth decade of life. Malignant TS are
very infrequently reported and mainly occur associated with neurofibromatosis type
1.[7]
The clinical presentation of these tumors depends on location and size; however, it
is common to see trigeminal hypesthesia, facial pain, and headaches, sometimes in
combination with hearing impairment, seizures, diplopia, ataxia, hemiparesis, papilledema,
jaw weakness, obstructive hydrocephalus, facial paralysis, and hemifacial spasm.[6]
[8]
[9]
[10]
[11]
[12]
Diagnosis is obtained from the clinical history, with magnetic resonance imaging (MRI)
as the imaging modality of choice. Characteristic lesional radiographic features are
a hypointense signal on T1, a hyperintense signal on T2, and avid contrast uptake
after intravenous gadolinium injection. Computed tomography (CT) scanning of the skull
is a complimentary exam aiding in the assessment of the skull base anatomy for surgical
planning.[2]
Material and Method
A retrospective analysis was performed of a cohort consisting of 14 patients with
TS treated at the Department of Neurosurgery of Santa Paula Hospital in São Paulo,
Brazil from January 1999 to October 2016. Our data were then compared with results
from published articles, after performing a systematic review and meta-analysis. A
PubMed search was done from October to December 2016, using the keywords “Brain Tumor”,
“Trigeminal Schwannoma”, “Microsurgery”, and “Radiosurgery.” We extracted relevant
data and created tables for a pooled analysis for comparison.
Classification
Lesions were classified according to 5 anatomically discernible segments of the nerve:
(1) intra-axial (brainstem), (2) cisternal, (3) Meckel's cave, (4) cavernous sinus,
and (5) skull base/extracranial. Tumors were further characterized according to their
composition into solid, cystic, or combined solid–cystic lesions.[2] They were subdivided according to size into small (< 3 cm), medium (3–4 cm), large
(4–5 cm), and giant (> 5 cm).[6] Another classification system was introduced by Samii and colleagues,[12] who classified tumors based on radiographic criteria into 4 types: Type A = intracranial
tumors, predominantly in the middle fossa; Type B = intracranial tumors, predominantly
in the posterior fossa; Type C = intracranial dumbbell-shaped tumors in the middle
and posterior fossa; Type D = extracranial tumors with intracranial extensions ([Figs. 1]
[2]
[3]
[4]).
Fig. 1 Type A, intracranial tumor predominantly in the middle fossa.
Fig. 2 Type B, intracranial tumor predominantly in the posterior fossa.
Fig. 3 Type C, intracranial dumbbell-shaped tumor in the middle and posterior fossa.
Fig. 4 Type D, extracranial tumor with intracranial extensions.
Results
During the study period, 14 adult patients were treated microsurgically. The cohort
consisted of 13 female and 1 male patients. The mean age of the cohort was 40 years
(range: 19–65 years). Based on size, 5 of the lesions were classified as medium in
size, 6 as large and 3 as a giant. Lesions were located on the left side in 9/14 patients
(64%) All patients were in good clinical status (Karnofsky Perfomance Scale [KPS] > 90)
with mild functional compromise only, but showed a high incidence of headaches and
vomiting ([Table 1]). The most frequent clinical symptom was headaches, followed by hypesthesia in 1
of the trigeminal branches (see [Table 2]).
Table 1
Case series
|
Case/gender
|
Age (y)
|
KPS (%)
|
Grade of resection
|
Follow-up
|
Side
|
Size (cm)
|
Classification
|
|
1/M
|
35
|
100
|
GTR
|
17 y
|
L
|
4.5
|
C
|
|
2/F
|
54
|
90
|
GTR
|
16 y
|
L
|
5
|
C
|
|
3/F
|
65
|
100
|
GTR
|
10 y
|
R
|
4.5
|
C
|
|
4/F
|
52
|
100
|
GTR
|
8 y
|
R
|
5
|
C
|
|
5/F
|
19
|
100
|
SR
|
8.5 y
|
L
|
4
|
C
|
|
6/F
|
34
|
100
|
GTR
|
7 y
|
R
|
4
|
A
|
|
7/F
|
43
|
100
|
GTR
|
6.5 y
|
L
|
6
|
C
|
|
8/F
|
32
|
100
|
SR
|
6 y
|
L
|
4.5
|
A
|
|
9/F
|
22
|
90
|
SR
|
6 y
|
R
|
6.5
|
C
|
|
10/F
|
28
|
100
|
SR
|
6 y
|
L
|
4
|
C
|
|
11/F
|
19
|
100
|
GTR
|
6 y
|
R
|
3
|
C
|
|
12/F
|
53
|
100
|
GTR
|
2 y
|
L
|
4.5
|
C
|
|
13/F
|
55
|
100
|
GTR
|
1 y
|
L
|
4
|
A
|
|
14/F
|
65
|
100
|
GTR
|
2 mo
|
L
|
6
|
C
|
Abbreviations: GTR, gross total resection; L, left; mo, month; R, right; SR, subtotal
resection; y, year.
Table 2
Signs and symptoms presented in the preoperative
|
Headache
|
50%
|
|
Vomit
|
21%
|
|
Behavior
|
14%
|
|
Strabismus
|
14%
|
|
Tinnitus
|
14%
|
|
Papilledema
|
21%
|
|
Hypoesthesia
|
28%
|
|
Paresis III
|
7%
|
|
Paresis V
|
14%
|
|
Paresis VII
|
14%
|
|
Pyramidal signs
|
7%
|
Tumors were predominantly cystic in 12/14 patients (85%) and had an average size of
4.6 cm (see [Fig. 5]). Three out of 14 patients (21%) presented with symptomatic hydrocephalus, requiring
the placement of a ventriculoperitoneal shunt. One of those patients, showed worsening
signs of intracranial hypertension during the follow-up period, while the other 2
had a resolution of their hydrocephalic symptoms after the shunting procedure.
Fig. 5 Magnetic resonance imaging (MRI) in axial sections with contrast, preoperative.
During the postoperative period, 1 patient developed wound infection, and 8 patients
showed some form of cranial nerve palsy and 2 patients had surgical wound hemorrhage.
None of these cases required a second operation, but was manageable expectantly.
Based on presurgical imaging, 3 patients of our cohort presented with type A lesions
(predominantly located in the middle fossa; 21% and 11 tumors were intracranial dumbbell-shaped
type C tumors, mainly located in the middle and posterior fossa (78%). Gross total
resection was achieved in 10 patients (71% of cases) based on intraoperative assessment
and postoperative MRI (see [Fig. 6]). Patients were regularly followed up after surgery with MRI and clinical visits,
with recurrences observed in 3 cases (2 after subtotal resection and 1 after an alleged
gross total resection). Further surgery was necessary for 1 of these 3 patients. All
the 3 patients with residual disease subsequently underwent stereotaxic radiosurgery,
gamma-knife, for adjuvant treatment.
Fig. 6 MRI, T1 with contrast, demonstrating gross subtotal resection (GSR).
Discussion
The introduction and improvement of available microscopic techniques and neuroanesthesia
have decreased the morbidity and mortality of patients undergoing surgery for trigeminal
schwannomas.[13] Over the years, complete resection with minimal morbidity and mortality has become
the goal of treatment with an emphasis on maintaining the quality of life in TS patients.
Complete resection is the most aggressive form of surgical treatment and offers potentially
a cure or prolonged progression-free survival, but is frequently associated with a
high incidence of neurological deficits.
Surgical
TS microsurgery aims at achieving a complete resection of the lesion and should be
performed without adding any new neurological deficit. However, incomplete resections
have been noted to be mostly related to inadequate exposure or complex tumor invasion
of the cavernous sinus or dense tumor adherence to the brainstem.[9]
[14]
[15]
[16] Due to improved microsurgical techniques, tumor recurrence rates are gradually decreasing
and range from 0 to 17%.[17] For tumors smaller than 3 cm, gross total resection rates of more than 95% have
been reported.[11] However, if tumor recurrence occurs, it usually does so in the cavernous sinus and
Meckel's cave.[17]
Surgical approaches were based on the classification of the lesion and are described
in [Table 1], while comparable results from the literature are shown in [Table 3]. Thus, for type A tumors, surgeons may use a pterional, orbitozygomatic, or subtemporal
approach. Type B tumors can be best approached via a retrosigmoid craniotomy (in particular,
if they are almost entirely in the posterior fossa). If a significant portion of the
tumor is located in the middle fossa, a transpetrous approach, or a combination of
approaches must be used. Type C tumors are the most frequent tumor type reported.
These TS are complex and usually require combined approaches to the middle fossa,
most commonly an anterior petrosectomy, and posterior fossa, as posterior petrosectomy,
and retrosigmoid approach.[18] Type D tumors can be accessed using a preauricular infratemporal approach.[5] Lesions that reach peripherally into the maxillary sinus and originate from the
middle fossa or cavernous sinus, can be accessed by a combination of transmaxillary
and middle fossa approaches.[6]
Table 3
Trigeminal schwannoma case series comparison
|
Author
|
Y
|
Radical removal (%)
|
Mortality (%)
|
Morbility (%)
|
Type A
|
Type B
|
Type C
|
Type D
|
Total
|
|
Yoshida and Kawase[18]
|
1999
|
20 (74%)
|
0
|
74
|
4
|
5
|
10
|
8
|
27
|
|
Goel et al[10]
|
2003
|
51 (70%)
|
3
|
7
|
29
|
7
|
30
|
7
|
73
|
|
Bhawani et al[3]
|
2006
|
35 (76%)
|
1
|
15
|
0
|
0
|
57
|
0
|
57
|
|
Pamir et al[23]
|
2007
|
17 (94%)
|
0
|
28
|
5
|
2
|
9
|
2
|
18
|
|
Fukaya et al[7]
|
2010
|
46 (81%)
|
2
|
68
|
15
|
12
|
26
|
4
|
57
|
|
Wanibuchi et al[16]
|
2012
|
86 (82%)
|
0
|
9
|
39
|
22
|
33
|
14
|
105
|
|
Chen et al[9]
|
2014
|
52 (95%)
|
0
|
5
|
13
|
10
|
21
|
11
|
55
|
|
Samii et al[17]
|
2014
|
15 (75%)
|
0
|
4
|
8
|
1
|
8
|
4
|
20
|
|
Jeong et al[21]
|
2014
|
47 (95.9%)
|
0
|
18
|
20
|
20
|
9
|
0
|
49
|
|
Current study
|
2016
|
10 (71%)
|
0
|
3
|
3
|
0
|
11
|
0
|
14
|
We believe that the best results can be achieved with small-sized lesions and tumor
configurations of types A or B. However, our series did not present with any lesion
classified as small (< 3cm), and only 3 type-A cases were encountered.
According to the literature, complications depend on size, location and approaches
chosen, but the most frequently reported postoperative complications were: meningitis,
fluid fistulas, masseter muscle atrophy, trigeminal pain and facial paresthesia, in
addition to paresis of the cranial nerves III–VI.[17]
Due to high surgical morbidity seen with TS resections, the field is progressively
evolving toward minimally invasive techniques.
Biopsy
Percutaneous biopsy is rarely employed, but offers a diagnostic alternative for lesions
in Meckel's Cave or lesions along the third division of the trigeminal nerve. The
biopsy can be performed employing a percutaneous technique comparable to approaches
used for trigeminal neuralgia[18]
[19]
[20] and can be done before any adjuvant treatment (e.g., SRS). We want to emphasize
the relevance of this approach as published by Sindou and colleagues, who reported
on a cohort of 50 patients, who underwent a percutaneous biopsy. Among these 50 cases,
which were biopsied due to insufficient imaging features to set the correct diagnosis,
the authors identified only the case 3 schwannomas exemplifying the importance of
this minor procedure for avoiding unnecessary open surgery, when other nonsurgical
pathologies are found.[20]
[21]
Endoscopic Endonasal
More recently, endoscopic transnasal surgical (ETS) approaches have been reported
as suitable for lesions affecting branches of V2 and V3. This technique allows accessing
tumors that are located paramedian, in the parasellar region, in Meckel's cave, as
well as lesions with discrete extension into the posterior fossa.[1] This technique can also be used for lesions in the middle fossa, pterygopalatine
fossa, and infratemporal fossa.[1]
[19] Arguments in favor of this approach are less retraction on temporal structures or
the cerebellum and good access to structures in the vicinity of the brainstem. However,
risk factors associated with ETS include a higher rate of CSF (cerebrospinal fluid)
liquorrhea and a potential risk of vascular complications due to the topographic relation
of the tumor to the internal carotid artery.[6]
Radiosurgery
Radiosurgery aims to control tumor growth without open surgical access and hence intends
to minimize any new treatment associated neurological deficits. To achieve this goal,
focal irradiation is delivered on the lesion,[22]
[23] with the aim to arrest further growth and possibly induce tumor necrosis and shrinking
of the mass. Radiosensitivity of the tumor depends on a variety of factors, including
its composition. Less fibrous tumors with higher cellularity and with cystic areas
are better candidates for gamma-knife surgery.[11] Some studies have shown an average decrease of TS size by approximately 6.8 mL,
over a follow-up of a period of 44 months.[4]
[22] In our series, 3 young patients, who received a partial resection due to complications
risk, were treated with gamma-knife surgery, once they presented MIB-1 above 3%, and
so they had a higher risk of regrowth and possibly the need of a new posterior surgery.[2]
[24]
The advancement of modern neuro-oncology increasingly relies on molecular studies,
obtained from the surgical specimen. When noninvasive SRS treatment is chosen empirically,
the indication is based on a clinical and radiological diagnosis alone, which carries
the risk of a misdiagnosis, since the differential diagnoses for tumors in this location
that includes other entities, such as meningiomas, epidermoid cysts, metastasis, chondrosarcomas,
chordomas, chondromas, other schwannomas, and maxillary sinus tumors. One needs to
be aware of the fact that such treatment without available histopathology may thus
occasionally lead to inadequate initial treatment or delay in obtaining the correct
diagnosis.[21]
[25]
Albeit rare, another problem is posed by malignant trigeminal schwannomas, which may
present with the local disease, as well as systemic hematogenous dissemination to
the lung and bone in up to one-third of patients.[7] SRS is a local treatment and therefore, unsuitable to address this potentially life-threatening
diagnosis. One has to remain vigilant about this diagnosis and potential metastases
during follow-up.
Conclusion
In this study, we report our experience of microsurgical treatment in a sizeable cohort
of patients with trigeminal schwannomas and conclude that the best treatment for large
symptomatic lesions remains open surgery. It allows for a gross total resection and
provides the patient with immediate relief of symptoms of mass effect. It offers the
patient a surgical cure and if exposure is optimal can often be performed without
adding significant new neurological deficits. In cases of subtotal resection (STR),
radiosurgery should be added to achieve a prolonged progression-free survival. Gentle
STR in conjunction with SRS may also result in lower overall treatment morbidity and
allow for a better quality of life posttreatment. Satisfactory results have also been
achieved with primary radiosurgery for smaller lesions (less than 30 mm), as well
as for lesions that are unresectable.
