Professional Documents
Culture Documents
- BACKGROUND: Trigeminal neuralgia (TN) affects 7% of was satisfactory and, with our dose/volume constraints, no
patients with multiple sclerosis (MS). In such patients, TN sensory complications were recorded. Nonetheless, long-
is difficult to manage either pharmacologically and surgi- term pain control was possible in less than half of the
cally. Radiosurgical rhizotomy is an effective treatment patients. This is a limitation that CyberKnife radiosurgery
option. The nonisocentric geometry of radiation beams of shares with other techniques in MS patients.
CyberKnife introduces new concepts in the treatment of
TN. Its efficacy for MS-related TN has not yet been
demonstrated.
- METHODS: Twenty-seven patients with refractory TN
INTRODUCTION
and MS were treated. A nonisocentric beams distribution
was chosen; the maximal target dose was 72.5 Gy. The
maximal dose to the brainstem was <12 Gy. Effects on pain,
medications, sensory disturbance, rate, and time of pain
recurrence were analyzed.
T rigeminal neuralgia (TN) is the most common craniofacial
pain syndrome, with an incidence of up to 5 in 100,000. It
is a severe condition requiring long-term medical treat-
ment. Nonetheless, up to 10% of patients suffer major adverse
drug-related events and require some type of surgical treatment.1,2
- RESULTS: Median follow-up was 37 (18e72) months. About 1%e2% of TN cases are caused by demyelinating plaques of
multiple sclerosis (MS) along the trigeminal pathway, nerve, and
Barrow Neurological Institute pain scale score IeIII was
brainstem. Trigeminal pain affects up to 7% of patients with MS,
achieved in 23/27 patients (85%) within 45 days. Prescrip-
and symptoms are often atypical or bilateral.3 In such patients, TN
tion isodose line (80%) accounting for a dose of 58 Gy is often difcult to manage either pharmacologically or surgically,
incorporated an average of 4.85 mm (4e6 mm) of the nerve with lower response rates than idiopathic TN.4-6
and mean nerve volume of 26.4 mm3 (range 20e38 mm3). Pioneered by Lars Leksell in 1951,7 stereotactic radiosurgery is a
Seven out of 27 patients (26%) had mild, not bothersome, proven and valuable method to treat TN. A remarkable body of
facial numbness (Barrow Neurological Institute numbness experience is available in the use of Gamma Knife single
score II). The rate of pain control decreased progressively isocenter treatments of TN.8-16 On the other hand, only a hand-
after the first year, and only 44% of patients retained pain ful of dedicated studies about the treatment of MS-related TN are
control 4 years later. available to date.17-23 Whether the radiosurgical rhizotomy for TN
can be performed using a frameless technique is often questioned.
- CONCLUSIONS: Frameless radiosurgery can be effec- The CyberKnife (Accuray Inc., Sunnyvale, California, USA), a
tively used to perform retrogasserian rhizotomy. Pain relief frameless robotic system,24-26 has been proposed for the treatment
Clinical Characteristics
Table 1 summarizes demographic and preoperative clinical data.
reconstruction increments of 0.75 mm, lter reconstruction H31
Preoperatively, all patients had severe pain with a numerical
(smooth), and 512 512 matrix.
rating scale score of 10 and were in Barrow Neurological
MRI was performed with the following parameters: 1) T1 volu-
Institute (BNI) class IV (33%) or V (67%). Twenty-two out of 27
metric sequence: matrix 512 512, ip angle 0 degrees, effective
patients (81%) had typical (TN1) trigeminal neuralgia. Fifty-nine
thickness 0.88 mm, reconstruction slice 0.75 mm, and recon-
percent had V3 involvement (22% V3 alone, 33% V2 and V3, 4%
struction increment 0 mm; 2) constructive interference in steady
V1-V2-V3); 88% had V2 involvement (22% V2 alone; 33% V2 and
state (CISS) sequence: matrix 512 512, ip angle 60, and
V3; 19% V1 and V2; 4% V1-V2-V3); and 23% had V1 involvement
reconstruction voxel size 0.35 mm; and 3) 3-dimensional time-of-
(19% V1 and V2; 4% V1-V2-V3). Pain was referred to the left side in
ight: matrix 512 512, ip angle 60, and reconstruction voxel
39% of patients. All patients had taken pain medication for an
size 0.39 mm. The axial source images were transferred to the
average of 2.3 years (range, 11 monthse7 years). One third of
CyberKnife workstation for treatment planning (Multiplan
patients had undergone previous rhizotomy (radiosurgery or
System).
radiofrequency).
Contouring of the target and critical volumes was dened on
Five patients had visible signal anomalies on T2-weighted se-
coregistered MRI and CT. The 5- to 6-mm retrogasserian segment
quences in the brainstem. Seven patients had a neurovascular
of the trigeminal nerve was used as a target. This volume was
contact, visible on magnetic resonance imaging (MRI), at the root
delineated on the CISS sequence with the aid of the CT. On this
entry zone (REZ) on the symptomatic side.
latter examination, it was possible to clearly identify the bony
landmarks of the nerve entrance in the Meckel cave (Figure 1).
Radiosurgery Technique
Immobilization and Imaging. Patients were treated with SRS using a
CyberKnife G4 model. Before the treatment, the patient lay supine Target and Doses Selection. Two surgeons (AC and AP) coregistered
on the treatment couch and a custom-tted thermoplastic mask the CT and MRI datasets and checked the quality of coregistration
(Ort Industries America, Jericho, New York, USA) was molded. visually using multiple views and transparency tools of the treat-
For all patients, a multislice computed tomography (CT) scan ment planning system (Multiplan, Accuray Inc.) in the 3 pro-
(Siemens Somatom Sensation 16, Siemens AG Medical Solutions, jections. Afterwards, they identied the Gasser ganglion and
Erlangen, Germany) and gadolinium-enhanced MRI (Siemens retrogasserian portion of the trigeminal nerve on the MRI and a
Magnetom 1.5-T) were performed. The CT protocol followed the bony canal on the edge of the petrous bone clearly identifying the
CyberKnife-specic requirements, namely, acquisition 16 0.75 entrance of the Meckel cave. This point could be immediately and
mm, Kv 120, effective mAs 320, rotation time 1 second, pitch 1.15. constantly identiable on 0.75-mm thin CT slices in axial view and
We chose to have a reconstruction slice of 0.75 mm and then checked on the sagittal and coronal views using a crosshair.
Figure 1. The crosshair is positioned on the bony landmark that was used to bony landmark allowed the submillimetric precision in coregistration of the
localize the trigeminal nerve at the entrance in the Meckel cave. Switching stereotactic computed tomography used by the system for image guidance
on the magnetic resonance imaging (MRI), the crosshair is positioned on the and MRI.
retrogasserian portion of the nerve, namely on the target. The use of this
Shifting the imaging from CT to the CISS sequence, the pars isodose line (58 Gy). The maximal length and volume of the nerve
triangularis was pointed out by the crosshair (see Figure 1). An that were eventually included into the 80% isodose line were
elongated target, 6 mm long and including the lateral margins determined by individual anatomy (length of the nerve and the
of the nerve, was drawn on 2 or 3 slices depending on the nerve relative dose received by the brainstem and the mesial temporal
thickness. The brainstem, mesial temporal lobe, acoustic and lobe). For shorter nerves, we moved the target forward, toward the
facial nerves, cochlea, and semicircular channels were Gasser ganglion, but always remaining into the root. If this was
specically delineated as critical structures to minimize radiation not sufcient to respect the dose limit to the brainstem, we forced
dose with the inverse planning algorithm. Other critical volumes the inverse planning accepting a smaller portion of the nerve to be
including the eyes, lens, optic nerves, whole brain, and skin included into a 58-Gy isodose line. The treatment plan was per-
were also delineated for dose calculations. Furthermore, formed the 1e3 days after imaging acquisition and 1e3 days
2 tuning structures were delineated at 3-mm and 10-mm dis- before the treatment delivery. Overall, the planning procedure
tances to restrict isodose distribution outside the target within a required 45e90 minutes.
precise distance.
The treatment was planned using a trigeminal node set, a Patient Setup, Treatment Delivery, and Quality Assurance. Patients
specically dened path of the robot, including a reduced source- were set up on the treatment couch utilizing a mask that had been
axis distance, which provides an effective collimation diameter of custom formed at the time of simulation. In-room lasers dene
4 mm at the isocenter. the center of the imaging system and provide the radiation ther-
A nonisocentric beam distribution was chosen; the maximal apists with an estimate for patient initial alignment. For treatment
dose delivered to the target was set at 72.5 Gy. The maximal point tracking, the 6D Skull Tacking mode was used. The treatment
doses to the brainstem and medial temporal lobe were set at 12 Gy location system compares orthogonal kV x-ray pairs, so-called live
and 36 Gy, respectively. Eight Gy was the maximal dose allowed to images, obtained during the patient setup to the planning sys-
the cranial nerves, whereas 4 Gy was the limit to the middle ear. temegenerated digitally reconstructed radiographs obtained from
The external most tuning structure (10 mm outside the target) had the stereotactic CT scan. Adjustments were then made to x-ray
a dose limit of 14 Gy. energy, mA, and pulse time to improve the quality of the live
Once the calculation was obtained, we veried that a 4- to images. The patient was then aligned to within a few millimeters
6-mm segment of the trigeminal nerve was included in the 80% of the nal treatment location by a robotic couch. The system
recommends shifts and rotations including pitch, roll, and yaw, (GRAPHPAD, San Diego, California, USA) application was used.
which are conrmed and initiated by the therapists. This process Multivariate analysis was performed using the multiple logistic
continues iteratively until the residual offsets are within acceptable regression method. Variables that were statistically signicant in
values (<1 mm in translation and <0.5 degrees in rotation that are the univariate analysis were transformed into binary variables to be
within limits corrected by the robot during treatment). The used in the logistic regression model. To perform multivariate
treatment location system allows for an x-ray image and conr- analysis, the STATCALC 7.1.1 software (AcaStat, Poinciana, Flor-
mation of the patients position at a dened time intervals. For the ida, USA) was used. The values of P < 0.05 were considered
trigeminal neuralgia treatment, a frequency of 15 seconds was statistically signicant.
used for all patients to reduce intrafraction inaccuracy. Any re-
sidual offset between the patients simulation position and that at RESULTS
the time of treatment is accounted for through robotic positional
adjustments. At each beam position, the robot adjusts the target Target and Treatment Data
location according to the patient offsets. All treatments were performed in a single session. Median pre-
The accuracy of targeting of the 6D Skull Tacking mode was scription isodose line (80%) accounting for a dose of 58 Gy
veried by offsetting the anthropomorphic phantom used for the incorporated an average of 4.85 mm (4 to 6 mm) segment of the
End-to-End (E2E) test by a known amount within this range and trigeminal nerve, with a mean nerve volume of 26.4 mm3 (range
then delivering the plan radiation in this or several similar offset 20e38 mm3) (Figure 2). The median maximal dose was 72.5 Gy
positions. Such a test was performed by dedicated medical phys- (range, 71.8eGy74.4). Median number of beams was 105 (range
icists on a monthly basis. The median total treatment error (TTE) 90e110); median number of nodes was 87 (range 85e90).
of the 6D Skull Tacking mode was measured, using GAFchromic Treatment time ranged 45e55 minutes with beam-on time
lms for ballcube EBT3 phantom (Ashland Advanced Materials, ranging 15e21 minutes. The average new Conformation Index was
Bridgewater, New Jersey, USA), as 0.30 0.12 mm. 1.32 (range, 1.04e2.17); the median homogeneity index was 1.25.
Assessments of MRI quality were performed on a monthly basis
using special phantoms and included geometric accuracy, slice Pain Control and Sensory Dysfunctions
thickness and intervals accuracy, slice position accuracy, image The median follow-up time was 37 months (range, 18e72
intensity uniformity, uniformity of signal-to-noise ratio, percent months). Signicant pain relief (a decrease in numerical rating
signal ghosting, and low-contrast object detectability. Values of scale score of 5) was achieved in 23 out of 27 (85%) patients
spatial linearity or geometric distortion were represented as the within 45 days. Twenty patients (74%) experienced pain cessation
maximal distance between rods of a reticle in a special phantom. within 15 days after treatment.
The geometric distortion depends on the gradients linearity and Twenty-three patients (85%) were able to decrease the frequency
magnetic eld linearity. Median vertical deviation was measured of pain medication throughout follow-up and 7 patients (26%)
as 0.4 mm; median horizontal deviation was 0.2 mm. ceased pain medication. Twelve months after treatment, the rate
of patients with BNI score of II or III was 85%; at 24 months, it
Follow-Up and Assessment of Outcome was 65%; at 36 months, it was 55%; at 48 months, it was 44%.
Follow-up information was obtained by outpatient clinical evalu- Figure 3 shows the rate of pain control after treatment.
ation or telephone interviews. The end points analyzed were Seven out of 27 patients (26%) had mild, not bothersome facial
1) effects on pain scores, 2) effects on medication, 3) latency to numbness (BNI numbness score II); all these patients had previ-
pain reduction, 4) occurrence of sensory disturbance, and 5) rate ously received rhizotomy treatments (P < 0.001). No patient re-
and time of pain recurrence. Pain level was scored using the BNI ported bothersome numbness.
scale (class I: no trigeminal pain, no medication; II: occasional Statistical analysis showed that a shorter nerve length (4 mm vs.
pain, not requiring medication; IIIa: no pain, continued medica- 5e6 mm) and smaller nerve volume (<25 mm3 vs. 25 mm3) were
tion; IIIb: controlled with medication; IV: some pain, not associated with treatment failure (P 0.02 and P 0.004,
adequately controlled with medication; V: severe pain, no pain respectively). These variables did not retain statistical signicance
relief).29 For hypoesthesia evaluation, we used the BNI facial in the multiple logistic analysis.
hypoesthesia scale (class I: no facial numbness; II: mild facial Multiple dermatomes involved in the trigeminal pain (P 0.97),
numbness, not bothersome; III: facial numbness, somewhat left side (P 0.16), male gender (P 0.68), atypical pain
bothersome; IV: facial numbness, very bothersome).29 We also (P 0.55), previous surgery (P 0.9), ipsilateral neurovascular
recorded any possible trigeminal motor decits. conict (P 0.55), visible brainstem plaques (P 0.14), and post-
treatment hypoesthesia (P 0.43) were not signicantly associ-
Statistical Analysis ated with treatment failure.
First, a descriptive analysis of recorded data was realized among Kaplan-Meier analysis was performed to compare pain control
the MS-related TN population. For the evaluation of outcomes in patients with typical versus atypical neuralgia. The log-rank test
such as initial pain cessation and recurrence, time to event was did not disclose a signicant difference (z 0.82; P 0.41). Pa-
estimated by using the Kaplan-Meier method. A bivariate analysis tient outcomes are listed in Table 2.
was then performed to identify predictive factors among the
collected variables. Contingency tables (Fisher exact test) were Salvage Therapy
used to compare categorical variables in univariate analysis. Fifteen patients (66%) did not benet from or experience recur-
To perform univariate analysis, the INSTAT 3.0 software rence after treatment. The median time to recurrent pain was of
Figure 2. Frameless radiosurgery for trigeminal neuralgia treatment plan. outside the 10% isodose line. Cranial nerve VIII, the cochlea, and the inner
The 4e6 mm (20e40 mm3) retrogasserian section of the trigeminal nerve ear received doses far below conventional tolerance limits.
was targeted, excluding the root entry zone. The brainstem was kept
24 months (range 18e42). Because of medically refractory pain, 10 Overall, the studies specically addressing the use of radio-
patients (37%) required further surgeries. Five patients received surgery for TN cases associated with MS report clinically relevant
radiofrequency retrogasserian rhizotomy, 3 patients underwent benets in 57%e100% of the patients17,19-23 (Table 3). On the
retreatment with CyberKnife, and 2 patients underwent micro- other hand, pain recurrence remains the main limitation of
vascular decompression and open retrogasserian rhizotomy. All 10 radiosurgery for TN. In our series, pain control was retained in
patients obtained pain control after the second treatment (BNI only 44% of the cases 4 years after the treatment, which is not
pain scores IeIIIa). All 10 patients reported new facial numbness different from what was reported in the Gamma Knife
(8 BNI numbness score II; 2 BNI numbness score III). series.8-10,12,16,30 Noteworthy, no patient in our series developed
bothersome numbness as a consequence of treatment.
The use of frameless stereotactic radiosurgery for the treatment
DISCUSSION of TN was rst reported by Romanelli et al,31 at Stanford in a study
In our series, the rst on the use of frameless, nonisocentric that was the rst clinical demonstration of the submillimetric
radiosurgery technique to treat MS-related TN, we observed initial accuracy of frameless radiosurgery. Almost immediate pain relief
pain relief in 85% of the patients. This result is consistent with the (within days) was found in this rst cohort of patients following
data of Gamma Knife radiosurgery for idiopathic TN, in which delivery of a prescribed dose ranging from 65e70 Gy to a nerve
initial pain relief has been reported in 50%e96%.8-10,12,16,30 segment up to 11 mm. The irradiation of such a long nerve
Figure 3. Kaplan-Meier estimate of the probability of patients with typical (TN1) and atypical (TN2) trigeminal
pain relief after treatment (A). Comparison between neuralgia (B).
segment, however, caused a high rate of bothersome numbness Indeed, the treatment planning for CyberKnife radiosurgical
that developed over time by these patients and suggested a retrogasserian rhizotomy introduces a number of novel yet highly
reduction of doses and length of the nerve to be treated. critical treatment planning variables, particularly the necessity to
RF, radiofrequency; RS, radiosurgery; NRS, numerical rating scale; BNI, Barrow Neurological Institute.
identify a favorable proportion between the radiation dose and but 47% suffered new facial numbness.37 An update from the
target volume of trigeminal nerve. Stanford series reported on 46 patients receiving a treatment
It has been difcult to determine an optimal dose range for delivered over a 6-mm segment of the nerve, with a mean mar-
CyberKnife treatments32-36 (Table 4). In 2008 Villavicencio et al37 ginal prescription dose of 58.3 Gy and a mean maximal dose of 73.
published the results of a multicenter study illustrating the 5 Gy.24 Symptoms disappeared completely in 39 patients (85%)
results of 95 patients who underwent CyberKnife radiosurgery. after a mean latency of 5.2 weeks. In most of these patients,
This heterogeneous study included patients treated with pain relief began within the rst week. Pain recurred in
different modalities (isocentric and nonisocentric) and doses. 1 patient after a pain-free interval of 7 months and was resolved
The median dose used was 78 Gy. Certain variables were by a second treatment. Four additional patients repeated the
predictive of stable pain relief over pain recurrence including the treatment after failing to respond adequately to the rst operation.
median maximum dose (77.5 vs. 65 Gy), median minimum dose After a mean follow-up period of 14.7 months, patient-reported
(64 vs. 52 Gy), and median nerve length treated (4 mm vs. 6 outcomes were excellent in 33 patients (72%), good in 11 pa-
mm). After 2 years, 50% in the population had excellent results, tients (24%), and poor in 2 patients (4%). Nevertheless,
Recurrence
numbness scale) was reported in 7 patients (15%).24
14.3%
33.3%
37.8%
61.5%
14.3%
66%
The ability to dene the target volume on the basis of the in-
dividual patients anatomy remains a distinct advantage of the
CyberKnife. The previously mentioned studies show that Cyber-
Effects Knife radiosurgery to an elongated segment of the trigeminal
57.1%
5.4%
Side
13%
7%
0%
16%
39%
10%
nerve is associated with satisfactory pain control but still a sig-
nicant rate of patients suffering bothersome sensory distur-
bances. In our patients, the marginal dose was set to 58 Gy with a
Pain Control
97.3%
90.7%
>90%
80%
57%
88%
90%
85%
90.7%
100%
Pain
42%
35%
48%
No
56.7 (6e174)
67 (13e96)
37 (18e72)
(months)
17 (6e38)
Mean 37.6
39 (3e7)
tioned the nal target volume. We set the minimum target volume
Pars triangularis
Pars triangularis
Retrogasserian
REZ
REZ
REZ
much lower than reported all along the literature on the treatment
of TN,8-10,15,38-43 but we did set this limit considering that the dose
70e80 (22 pts), >80 (31 pts)
80 (5 pts), 90 (2 pts)
70e80 (12), >80 (3)
Median 90 (80e90)
59 (38e74)
62 (39e86)
53.3 (36e62)
51
27
15
37
13
43
35
124
related to the use of higher doses and a REZ target early in their
Diwanji et al., 201018
29
23
Rogers et al., 2002
to the brainstem at the REZ. Actually, the length of the nerve and
the consequent dose received by the pons remains the main
Authors
Table 4. Series Reporting Nonisocentric CyberKnife Radiosurgery for Treatment of Idiopathic Trigeminal Neuralgia
Results
range 5e12)
Villavicencio et al., 200837 95 69.8 78 Retrogasserian (median 6 mm; Mean 2 years 67% 92% 18% 28%
range 5e12)
Fariselli et al., 200932 33 74 55 (6 pts) Retrogasserian (4 mm) Mean 23 97% 0% 33%
65 (10 pts)
75 (17 pts)
FU, follow-up.
ORIGINAL ARTICLE
ORIGINAL ARTICLE
ALFREDO CONTI ET AL. FRAMELESS SRS FOR MS-RELATED TRIGEMINAL NEURALGIA
planning, all brain structures must be contoured. Automatic MS-related TN, the rate of recurrence ranged from 33%e61.5% in
contouring was used, and each volume was carefully checked and the long run.17,19-23 Unfortunately, other lesional techniques seem
corrected for individual anatomy. All dose volume histograms had to have even worse results in terms of long-term pain relief.
to be evaluated since the risk of high doses on sensitive structures, Recently Alvarez-Pinzon et al17 compared percutaneous
even though distant from the target, is remarkable because of the retrogasserian balloon compression and Gamma Knife
nonisocentric radiation beams distribution. In particular, the radiosurgery in 202 patients with MS. Fewer complications and
medialmost portion of the temporal lobe can receive doses as high superior long-term relief were associated with the Gamma
as 45 Gy after initial planning that was reduced to <15 Gy/1 mm3 Knife. Similarly, radiosurgery demonstrated fewer complications
to avoid late mediotemporal radionecrosis. than percutaneous retrogasserian glycerol rhizotomy for trigemi-
One of the main criticisms to the use of CyberKnife for TN is nal neuralgia in patients with multiple sclerosis.47 Therefore
the necessity to plan on CT cisternography or MRI after a poten- radiosurgery may represent the best treatment option for the
tially inadequate fusion. We suggest that bony landmarks, indi- treatment of TN in patients with MS, while other more invasive
cating the entrance of the trigeminal nerve root into the Meckel rhizotomy techniques should be reserved for patients with acute,
cave, are easily recognizable directly on a bone CT scan. The intractable pain requiring immediate postoperative relief.
identication of these points greatly supports a precise coregis- In conclusion, frameless radiosurgery represents a safe tech-
tration of the CT with magnetic resonance sequence, stressing the nique to achieve trigeminal retrogasserian rhizotomy. In MS pa-
T2 values between cerebrospinal uid and cranial nerves and tients, results are satisfactory, with 85% of patients gaining pain
vessels. Magnetic distortion is a possible limit of functional control within few weeks. Furthermore, using our constraints for
treatments performed by MRI targeting (i.e., we measured a dose, volume of the nerve, and dose to the brainstem, no both-
median maximal distortion of 0.4 mm). Nevertheless, this is a ersome hypoesthesia or other complications were recorded. In the
common limit of anatomic targeting in functional neurosurgery long term, however, pain control seems to be possible in less than
whose clinical impact appears negligible. half of the patients. This is a limit that radiosurgery shares with
The major limitation of our treatment modality seems to be the other retrogasserian rhizotomy techniques in MS patients. Higher
durability of pain relief, with only 44% retaining effective pain doses delivered to a longer nerve portion would be possible with
control 4 years after the treatment. It appears that this is common the CyberKnife resulting, according to the experience in patients
to frame-based isocentric radiosurgical rhizotomy. Actually, in with idiopathic TN, in long-term pain control but a higher risk of
Gamma Knife series specically addressing the treatment of sensory complications.
9. Marshall K, Chan MD, McCoy TP, Aubuchon AC, 16. Tuleasca C, Carron R, Resseguier N, Donnet A,
REFERENCES Bourland JD, McMullen KP, et al. Predictive vari- Roussel P, Gaudart J, et al. Patterns of pain-free
ables for the successful treatment of trigeminal response in 497 cases of classic trigeminal neu-
1. Bennetto L, Patel NK, Fuller G. Trigeminal neu-
neuralgia with Gamma Knife radiosurgery. ralgia treated with Gamma Knife surgery and
ralgia and its management. BMJ. 2007;334:
Neurosurgery. 2012;70:566-572. followed up for least 1 year. J Neurosurg. 2012;
201-205.
117(suppl):181-188.
10. Massager N, Lorenzoni J, Devriendt D,
2. Obermann M. Treatment options in trigeminal
Desmedt F, Brotchi J, Levivier M. Gamma Knife 17. Alvarez-Pinzon AM, Wolf AL, Swedberg HN,
neuralgia. Ther Adv Neurol Disord. 2010;3:107-115.
surgery for idiopathic trigeminal neuralgia per- Barkley KA, Cucalon J, Curia L, et al. Comparison
formed using a far-anterior cisternal target and a of percutaneous retrograsserian balloon
3. Burchiel KJ, Slavin KV. On the natural history of
high dose of radiation. J Neurosurg. 2004;100: compression and Gamma Knife radiosurgery for
trigeminal neuralgia. Neurosurgery. 2000;46:
597-605. the treatment of trigeminal neuralgia in multiple
152-154.
sclerosis: a clinical research study article. World
11. Pollock BE. Fitting radiosurgery into the trigemi- Neurosurg. 2017;97:590-594.
4. Broggi G, Ferroli P, Franzini A, Nazzi V, Farina L, nal neuralgia management puzzle. World Neurosurg.
La Mantia L, et al. Operative ndings and out- 2010;74:448-450. 18. Diwanji TP, Dhople AA, Chin LS, Maggio WW,
comes of microvascular decompression for tri- Adams JR, Regine WF, et al. Long-term outcome
geminal neuralgia in 35 patients affected by 12. Regis J, Metellus P, Hayashi M, Roussel P, of Gamma Knife stereotactic radiosurgery for
multiple sclerosis. Neurosurgery. 2004;55:830-838. Donnet A, Bille-Turc F. Prospective controlled trial multiple sclerosis associated trigeminal neuralgia.
of Gamma Knife surgery for essential trigeminal Int J Radiat Oncol Biol Phys. 2010;78:S264-S265.
5. Kanpolat Y, Berk C, Savas A, Bekar A. Percuta- neuralgia. J Neurosurg. 2006;104:913-924.
neous controlled radiofrequency rhizotomy in the 19. Huang E, Teh BS, Zeck O, Woo SY, Lu HH,
management of patients with trigeminal neuralgia 13. Riesenburger RI, Hwang SW, Schirmer CM, Chiu JK, et al. Gamma Knife radiosurgery for
due to multiple sclerosis. Acta Neurochir (Wien). Zerris V, Wu JK, Mahn K, et al. Outcomes treatment of trigeminal neuralgia in multiple
2000;142:685-689. following single-treatment Gamma Knife surgery sclerosis patients. Stereotact Funct Neurosurg. 2002;
for trigeminal neuralgia with a minimum 3-year 79:44-50.
6. Kondziolka D, Lunsford LD, Bissonette DJ. Long- follow-up. J Neurosurg. 2010;112:766-771.
term results after glycerol rhizotomy for multiple 20. Rogers CL, Shetter AG, Ponce FA, Fiedler JA,
sclerosis-related trigeminal neuralgia. Can J Neurol 14. Sheehan J, Pan HC, Stroila M, Steiner L. Gamma Smith KA, Speiser BL. Gamma Knife radiosurgery
Sci. 1994;21:137-140. Knife surgery for trigeminal neuralgia: outcomes for trigeminal neuralgia associated with multiple
and prognostic factors. J Neurosurg. 2005;102: sclerosis. J Neurosurg. 2002;97(suppl 5):529-532.
7. Leksell L. Sterotaxic radiosurgery in trigeminal 434-441.
neuralgia. Acta Chir Scand. 1971;137:311-314. 21. Tuleasca C, Carron R, Resseguier N, Donnet A,
15. Tawk RG, Duffy-Fronckowiak M, Scott BE, Roussel P, Gaudart J, et al. Multiple sclerosis-
8. Kondziolka D, Zorro O, Lobato-Polo J, Kano H, Alberico RA, Diaz AZ, Podgorsak MB, et al. Ste- related trigeminal neuralgia: a prospective series
Flannery TJ, Flickinger JC, et al. Gamma Knife reotactic Gamma Knife surgery for trigeminal of 43 patients treated with Gamma Knife surgery
stereotactic radiosurgery for idiopathic trigeminal neuralgia: detailed analysis of treatment response. with more than one year of follow-up. Stereotact
neuralgia. J Neurosurg. 2010;112:758-765. J Neurosurg. 2005;102:442-449. Funct Neurosurg. 2014;92:203-210.
22. Weller M, Marshall K, Lovato JF, Bourland JD, 32. Fariselli L, Marras C, De Santis M, Marchetti M, 42. Pollock BE, Phuong LK, Foote RL, Stafford SL,
deGuzman AF, Munley MT, et al. Single-institu- Milanesi I, Broggi G. CyberKnife radiosurgery as a Gorman DA. High-dose trigeminal neuralgia
tion retrospective series of Gamma Knife rst treatment for idiopathic trigeminal neuralgia. radiosurgery associated with increased risk of
radiosurgery in the treatment of multiple Neurosurgery. 2009;64(suppl 2):A96-A101. trigeminal nerve dysfunction. Neurosurgery. 2001;
sclerosis-related trigeminal neuralgia: factors that 49:58-62.
predict efcacy. Stereotact Funct Neurosurg. 2014;92: 33. Karam SD, Tai A, Snider JW, Bhatia S, Bedrick EJ,
53-58. Rashid A, et al. Refractory trigeminal neuralgia 43. Smith ZA, Gorgulho AA, Bezrukiy N, McArthur D,
treatment outcomes following CyberKnife radio- Agazaryan N, Selch MT, et al. Dedicated linear
23. Zorro O, Lobato-Polo J, Kano H, Flickinger JC, surgery. Radiat Oncol. 2014;9:257. accelerator radiosurgery for trigeminal neuralgia:
Lunsford LD, Kondziolka D. Gamma Knife radi- a single-center experience in 179 patients with
osurgery for multiple sclerosis-related trigeminal 34. Lim M, Cotrutz C, Romanelli P, Schaal D, Gibbs I, varied dose prescriptions and treatment plans. Int
neuralgia. Neurology. 2009;73:1149-1154. Chang SD, et al. Stereotactic radiosurgery using J Radiat Oncol Biol Phys. 2011;81:225-231.
CT cisternography and non-isocentric planning
24. Adler JR Jr, Bower R, Gupta G, Lim M, Efron A, for the treatment of trigeminal neuralgia. Comput 44. Aubuchon AC, Chan MD, Lovato JF,
Gibbs IC, et al. Nonisocentric radiosurgical rhi- Aided Surg. 2006;11:11-20. Balamucki CJ, Ellis TL, Tatter SB, et al. Repeat
zotomy for trigeminal neuralgia. Neurosurgery. Gamma Knife radiosurgery for trigeminal neu-
2009;64(suppl 2):A84-A90. 35. Singh R, Davis J, Sharma S. Stereotactic radio- ralgia. Int J Radiat Oncol Biol Phys. 2011;81:
surgery for trigeminal neuralgia: a retrospective 1059-1065.
25. Conti A, Pontoriero A, Arpa D, Siragusa C, multi-institutional examination of treatment out-
Tomasello C, Romanelli P, et al. Efcacy and comes. Cureus. 2016;8:e554. 45. Conti A, Pontoriero A, Siddi F, Iati G, Cardali S,
toxicity of CyberKnife re-irradiation and dose Angileri FF, et al. Post-treatment edema after
dense temozolomide for recurrent gliomas. Acta 36. Tang CT, Chang SD, Tseng KY, Liu MY, Ju DT. meningioma radiosurgery is a predictable
Neurochir (Wien). 2012;154:203-209. CyberKnife stereotactic radiosurgical rhizotomy complication. Cureus. 2016;8:e605.
for refractory trigeminal neuralgia. J Clin Neurosci.
26. Conti A, Pontoriero A, Ricciardi GK, Granata F, 46. Lowell D, Tatter SB, Bourland JD, deGuzman AF,
2011;18:1449-1453.
Vinci S, Angileri FF, et al. Integration of func- Ekstrand KE, Ellis TL, et al. Toxicity of Gamma
tional neuroimaging in CyberKnife radiosurgery: Knife radiosurgery in the treatment of intracranial
37. Villavicencio AT, Lim M, Burneikiene S,
feasibility and dosimetric results. Neurosurg Focus. tumors in patients with collagen vascular diseases
Romanelli P, Adler JR, McNeely L, et al. Cyber-
2013;34:E5. or multiple sclerosis. Int J Radiat Oncol Biol Phys.
Knife radiosurgery for trigeminal neuralgia treat-
2011;81:e519-e524.
ment: a preliminary multicenter experience.
27. Lipton RB, Bigal ME, Steiner TJ, Silberstein SD,
Neurosurgery. 2008;62:647-655.
Olesen J. Classication of primary headaches. 47. Mathieu D, Effendi K, Blanchard J, Seguin M.
Neurology. 2004;63:427-435. Comparative study of Gamma Knife surgery and
38. Fountas KN, Smith JR, Lee GP, Jenkins PD,
percutaneous retrogasserian glycerol rhizotomy
28. Eller JL, Raslan AM, Burchiel KJ. Trigeminal Cantrell RR, Sheils WC. Gamma Knife stereotac-
for trigeminal neuralgia in patients with multiple
neuralgia: denition and classication. Neurosurg tic radiosurgical treatment of idiopathic trigemi-
sclerosis. J Neurosurg. 2012;117(suppl):175-180.
Focus. 2005;18:E3. nal neuralgia: long-term outcome and
complications. Neurosurg Focus. 2007;23:E8.
29. Rogers CL, Shetter AG, Fiedler JA, Smith KA,
Han PP, Speiser BL. Gamma Knife radiosurgery 39. Jawahar A, Wadhwa R, Berk C, Caldito G,
for trigeminal neuralgia: the initial experience of DeLaune A, Ampil F, et al. Assessment of pain
The Barrow Neurological Institute. Int J Radiat control, quality of life, and predictors of success Conflict of interest statement: This research did not receive
Oncol Biol Phys. 2000;47:1013-1019. after Gamma Knife surgery for the treatment of any specific grant from funding agencies in the public,
trigeminal neuralgia. Neurosurg Focus. 2005;18:E8. commercial, or not-for-profit sectors.
30. Han JH, Kim DG, Chung HT, Paek SH, Kim YH,
Kim CY, et al. Long-term outcome of Gamma 40. Maesawa S, Salame C, Flickinger JC, Pirris S, Received 15 January 2017; accepted 17 April 2017
Knife radiosurgery for treatment of typical tri- Kondziolka D, Lunsford LD. Clinical outcomes Citation: World Neurosurg. (2017) 103:702-712.
geminal neuralgia. Int J Radiat Oncol Biol Phys. after stereotactic radiosurgery for idiopathic tri- http://dx.doi.org/10.1016/j.wneu.2017.04.102
2009;75:822-827. geminal neuralgia. J Neurosurg. 2001;94:14-20.
Journal homepage: www.WORLDNEUROSURGERY.org
31. Romanelli P, Heit G, Chang SD, Martin D, 41. Park SH, Hwang SK. Outcomes of Gamma Knife Available online: www.sciencedirect.com
Pham C, Adler J. Cyberknife radiosurgery for tri- radiosurgery for trigeminal neuralgia after a min-
1878-8750/$ - see front matter 2017 Elsevier Inc. All
geminal neuralgia. Stereotact Funct Neurosurg. 2003; imum 3-year follow-up. J Clin Neurosci. 2011;18:
rights reserved.
81:105-109. 645-648.