Docstoc

Repeat Gamma Knife surgery for trigeminal neuralgia long term results

Document Sample
Repeat Gamma Knife surgery for trigeminal neuralgia long term results Powered By Docstoc
					                                                                                                           J Neurosurg 113:178–183, 2010



                      Repeat Gamma Knife surgery for trigeminal neuralgia:
                      long-term results
                      Clinical article
                      Brent Y. KimBall, m.D.,1 JeffreY m. SorenSon, m.D., 2 anD DaviD Cunningham, m.D. 2
                      1
                       Department of Neurosurgery, University of Tennessee Health Science Center;
                      and 2Department of Neurosurgery, University of Tennessee and the Semmes-Murphey Neurologic
                      & Spine Institute, Memphis, Tennessee

                            Object. The purpose of this study was to assess the long-term outcome achieved after repeat Gamma Knife sur-
                      gery (GKS) for trigeminal neuralgia (TN) using a uniform treatment plan.
                            Methods. Between 1985 and 2010, 53 patients underwent repeat GKS for refractory TN. In the initial GKS,
                      which involved targeting the root entry zone of the trigeminal nerve, a maximal dose of 80 Gy was used with a 4-mm
                      collimator so that the 50% isodose line abutted the pons. In the second GKS, the treatment plan consisted of a 70-Gy
                      dose directed at a target 4–5 mm distal to the first target on the trigeminal nerve. The mean follow-up duration in these
                      patients was 42 months. Outcomes were defined using the Marseille scale: excellent (Class I or II, no pain with or
                      without medications), good (Class III or IV, ≥ 50% relief), and poor (Class V, < 50% relief).
                            Results. Trigeminal neuralgia pain was controlled (≥ 50% improvement with or without medications) after
                      repeat GKS in 70% of patients at 1 year, 50% at 3 years, 50% at 5 years, and 50% at 10 years, as defined by a Kaplan-
                      Meier analysis. A correlation was found between facial numbness and pain relief (p = 0.047). No difference was
                      found between patients with Type 1 TN and those with Type 2 TN, and there was no correlation between the best
                      relief obtained and long-term durability of relief from pain. Twenty-two patients (47.8%) described their trigeminal
                      dysfunction in the following manner: numbness (45.6%), dry eye (10.9%), taste change (8.7%), or jaw weakness
                      (2.2%). In only 8.7% of cases did the patient experience facial numbness that was regarded as bothersome.
                            Conclusions. Repeat GKS for TN at the doses used provides substantial long-term relief. Treatment failure oc-
                      curred up to 28 months after the second GKS. Facial numbness correlated with more durable pain relief after repeat
                      GKS in this series. (DOI: 10.3171/2010.8.GKS101075)

                      KeY WorDS      •      Gamma Knife surgery      •      trigeminal neuralgia      •
                      repeated treatment      •      pain



A
        lthough   an early application for GKS, TN was                    pain and treatment success, lack of long-term follow-up,
        not widely treated by this modality until MR                      and variations in patient selection criteria. In this article,
        imaging could be used to accurately target the                    we report our experience with second GKS for TN and
trigeminal nerve. In the interim, MVD became the defini-                  compare our results with those in the literature.
tive procedure to treat medically intractable TN, because
the majority of cases were believed to be due to vascu-
lar compression at the trigeminal nerve root entry zone.                                              Methods
Although excellent results have been shown with MVD,                           Between January 1985 and March 2010, 435 Gamma
there remains a role for less invasive alternatives, such                 Knife procedures were performed in 379 patients with
as GKS and percutaneous rhizotomy, both as initial and                    TN; 53 of these were repeat procedures on the same
salvage procedures. Unlike MVD, these ablative proce-                     nerve. All patients had facial pain that was resistant to
dures require a dosing decision that must strike a balance                medical management with typical agents such as pheny-
between pain relief and nerve dysfunction. Thus, in the                   toin, carbamazepine, gabapentin, and baclofen. Only pa-
context of a second GKS, it is important to learn as much                 tients who experienced significant improvement in their
as possible from clinical series in which a variety of dos-               pain after the initial GKS were offered repeat treatment.
es and treatment plans have been used, even though these                  Eight patients were excluded from the analysis due to a
studies are still challenged by unsettled definitions of                  lack of sufficient follow-up. One patient underwent repeat
                                                                          GKS on both nerves and was therefore classified as 2
   Abbreviations used in this paper: BNI = Barrow Neurological
                                                                          separate cases. A total of 46 nerves were treated in 45 pa-
Institute; GKS = Gamma Knife surgery; MVD = microvascular                 tients. Eight patients had previously undergone invasive
decompression; PBC = percutaneous balloon compression; PGR =              procedures: 1 had undergone RFL twice; 5 patients had
percutaneous glycerol rhizotomy; RFL = radiofrequency lesioning;          received prior MVD (including 1 patient who acquired
TN = trigeminal neuralgia.                                                facial pain after a tumor resection), 1 patient with MS un-

178                                                                                        J Neurosurg / Volume 113 / December 2010
Repeat GKS for trigeminal neuralgia: long-term results

derwent PGR 6 times prior to GKS, and 1 patient under-         TABLE 1: Patient characteristics in 46 cases of repeated GKS
went an infraorbital nerve avulsion. Two patients carried      for TN*
a diagnosis of MS, and another patient had a diagnosis of
systemic lupus erythematosus.                                                     Variable                       No. of Cases†
     In this series, 41% of patients were men, and 46% of       male sex                                        19 (41%)
treated nerves were on the left side. The mean patient age
                                                                mean age in yrs (range)                         68 (30–87)
was 68 years (range 30–87 years), and the median dura-
tion of symptoms prior to the first GKS was 4 years. The        median pain duration before 1st GKS (range)   4 yrs (3 mos–36 yrs)
mean interval between the first and second GKS was 33           mean interval btwn GKS treatments (range)       33 (3–138) mos
months. The most common pattern of pain was in both             type of pain
the V2 and V3 distributions of the trigeminal nerve in             1                                            35 (76%)
45.7% of cases, followed by the V2 alone in 21.7%, the V3          2                                            11 (24%)
alone in 19.6%, all trigeminal distributions in 6.5%, the          rt side pain                                 54%
V1 and V2 in 4.3%, and the V1 alone in 2.2% of cases.           pain distribution
The preoperative characteristics of our patients are sum-          V2 & V3                                      45.7%
marized in Table 1.
                                                                   V2                                           21.7%
     Although all patients complained of lancinating TN
pain on initial presentation, several also reported that           V3                                           19.6%
they experienced constant aching and throbbing pain as             V1 & V2 & V3                                  6.5%
well. Following the classification of Burchiel and Slavin,7        V1 & V2                                       4.3%
11 patients had Type 2 pain, which was described as con-           V1                                            2.2%
stant aching, burning, or throbbing more than 50% of the        prior invasive surgical intervention             8 (17.4%)
time. Thirty-five patients had Type 1 pain, which was de-
scribed as primarily sharp, lancinating, and/or shock-like     *  Forty-five patients underwent repeated GKS. One patient underwent 
pain with pain-free intervals.17                               repeat GKS on both trigeminal nerves and was therefore classified as 
                                                               2 cases.
Radiosurgical Technique                                        † Unless otherwise indicated.
     Gamma Knife models U and 4C (Elekta AB) were
used during this 25-year experience. After a local anes-
thetic agent had been given to each patient, a model G         were insufficient follow-up data in 8 patients, and their
stereotactic frame (Elekta AB) was affixed to the head,        cases were excluded from the outcome analysis. During
and all 53 patients underwent stereotactic MR imaging to       interviews, patients were asked about their degree of pain
identify the trigeminal nerve. Magnetic resonance imag-        relief, duration of pain relief, use of medications, need
ing was performed using contrast-enhanced, T1-weighted         for further surgical procedures, and the development of
axial phase-volume acquisitions with 1-mm slices. All          trigeminal dysfunction. The mean follow-up duration was
collimators were left open, and a single 4-mm isocenter        42 months (range 1–122 months).
was used in all patients for both the first and second GKS          Pain outcome was determined using a modified Mar-
treatments. The isocenter was placed at the trigeminal         seille scale (Table 2).22 Initial response was defined as the
nerve root entry zone with the 50% isodose line abutting       best pain relief shortly after the second GKS. Adequate
the pons. Before placing the retreatment target, a phan-       pain relief was defined as Marseille Scale Classes I–IV,
tom dose was created to replicate the initial treatment tar-   whereas Marseille Class V or additional surgical inter-
get. The retreatment isocenter was then placed 4–5 mm          vention defined treatment failure. After GKS, trigeminal
distal to the initial target, with the 50% isodose lines of    nerve sensory dysfunction was categorized as either both-
the targets less than 2 mm apart. This treatment plan was      ersome or mild by the patient. Patients with numbness
consistent in all but 1 case in which the target locations     were asked whether their numbness was a good trade-off
were reversed. The primary and secondary doses were            for the pain relief they received.
80 and 70 Gy, respectively, with a cumulative dose of 150
Gy, except for 1 patient who received a total dose of 140      Statistical Analysis
Gy and another who received a total dose of 160 Gy.                 The “R” statistical software package (http://www.
Outcome Assessment
                                                               r-project.org) was used for analyses. To evaluate the long-
                                                               term effectiveness of repeat GKS, we calculated survival
     This retrospective study was approved by the Univer-      curves using the Kaplan-Meier product-limit method. The
sity of Tennessee Institutional Review Board, and patient      log-rank test was used to determine statistical differences
informed consent was obtained. Long-term follow-up             between survival curves (p < 0.05). The relationships of
data from the patients’ medical records were supplement-       various clinical parameters with outcome were explored
ed by patient surveys completed via telephone interviews       (including patient sex and age, duration of symptoms be-
in 40 cases. Four patients with adequate medical records       fore the first GKS, interval between the first and second
could not be contacted for an interview. In 2 deceased         GKS, presence of numbness after the second GKS, type
patients, data were obtained through medical records and       of pain, and presence or absence of previous failed surgi-
telephone interviews with close family members. There          cal procedures).

J Neurosurg / Volume 113 / December 2010                                                                                       179
                                                                       B. Y. Kimball, J. M. Sorenson, and D. Cunningham
TABLE 2: Pain assessment scale                                             TABLE 3: Treatment outcomes in 46 cases

                                               Modified Marseille                          Parameter                      No. of Cases
   Outcome           Modified BNI Scale              Scale
                                                                               best response after 2nd GKS (Mar-
excellent       I, no pain & no medication    I, no pain & no medi-                seille class)
                    required                      cations                          I                                       21 (45.6%)
                II, occasional pain & no      II, no pain w/ medica-               II                                      11 (23.9%)
                    medications                   tions                            III                                      5 (10.9%)
good            IIIa, no pain but continued   III, ≥90% pain relief                IV                                       5 (10.9%)
                    use of medications re-                                            V                                     4 (8.7%)
                    quired                                                     Marseille class at last follow-up
                IIIb, some pain adequately    IV, ≥50% relief                    I or II (excellent)                      20 (43.5%)
                  controlled w/ medications                                      III or IV (good)                          6 (13.0%)
                                                                                   V or further procedures (poor)         20 (43.5%)
poor (treatment IV, pain improved but not     V, <50% relief
  failure)        controlled w/ medication                                     additional treatments after 2nd GKS
                                                                                   MVD                                         7
                                                                                   PGR                                         2
                                                                                 PBC                                           2
                            Results                                                MVD + PGR                                   1
                                                                                   RFL                                         1
     Overall, 91.3% of patients had a good response (≥                         trigeminal dysfunction
50% relief) after their second GKS procedure (Table 3).                            absent                                 24 (52.2%)
Sixteen (38%) of the initial responders experienced pain                           present                                22 (47.8%)
recurrence at the last follow-up. The degree to which a                               numbness                            21 (45.6%)
patient initially responded to their second GKS did not                               bothersome                           4 (8.7%)
significantly influence the treatment failure rate.                                   dry eye                              5 (10.9%)
     At the last follow-up, excellent results (Marseille                              taste change                         4 (8.7%)
Classes I and II) were maintained in 20 cases (43.5%),                                jaw weakness                         1 (2.2%)
good results (Marseille Classes III and IV) in 6 cases                                anesthesia dolorosa                      0
(13%), and poor results (Marseille V or additional sur-
gical treatment) in 20 cases (43.5%). The majority of
treatment failures occurred before 18 months (mean 8.5                                                 Discussion
months), and the last case of treatment failure occurred
at 28 months after the repeat GKS. Of the 20 patients in                   Pain Relief
whom repeat GKS ultimately failed, 13 underwent sub-                            Microvascular decompression is the gold standard
sequent surgical procedures: 7 patients were treated by                    for medically refractory TN, providing excellent relief
MVD, 2 patients by PGR, 2 patients by PBC, 1 patient by                    (Marseille Class I) in 75% of cases at 1 year and 64%
both MVD and PGR, and 1 patient by RFL. The results                        at 10 years.1 Kondziolka et al.15 demonstrated long-term
of these additional procedures were not statistically ana-                 durability of pain relief (≥ BNI Score IIIb) after the ini-
lyzed. Kaplan-Meier curves for adequate pain control (≥                    tial GKS at 1, 5, and 10 years to be 80%, 46%, and 30%,
50% relief) at 1, 3, 5, and 10 years were 70%, 50%, 50%,                   respectively. When GKS is used as the primary treat-
and 50%, respectively (Fig. 1).                                            ment for TN, complete pain relief with or without medi-
     No patient sustained an early complication after                      cation may be as high as 71% at 1 year and 66% at 4
GKS. Trigeminal dysfunction of any kind occurred in                        years post-GKS.13 Recent studies demonstrated that up to
22 patients (47.8%) and numbness in 21 patients (45.6%);                   89% of patients respond favorably to an initial GKS treat-
only 4 patients found this numbness to be bothersome.                      ment.15 Factors related to a more favorable outcome after
All patients with numbness believed it was a good trade-                   the initial GKS for TN include fewer prior surgical treat-
off for the accompanying pain relief (Table 4). Presence                   ments,6,16 lack of atypical features of pain,2,6,16 higher radi-
of facial numbness after the second GKS correlated with                    ation doses,20 and new trigeminal deficits after radiosur-
better long-term pain relief (p = 0.047). The 1-, 3-, 5-,                  gery.6,20,23 The results and risk factors of the second GKS
and 7-year Kaplan-Meier rates for adequate pain relief                     treatment are less well studied. Huang et al.13 analyzed
in patients with facial numbness were 84%, 64%, 64%,                       final outcomes in 89 patients treated primarily with GKS.
and 64%, respectively; the corresponding rates for pa-                     In that series, final outcomes after 1 or 2 GKS treatments
tients without numbness were 57%, 40%, 40%, and 40%,                       were as follows: excellent relief (no pain and no medica-
respectively. A univariate analysis found no significant                   tion) in 68.5%, good relief (no pain with medication) in
correlation between adequate pain relief and type of pain                  14.6%, and fair relief (≥ 50% relief) in 14.7% of patients.
(Type 1 vs Type 2) (p = 0.916), interval between proce-                    In total, ≥ 50% relief was found in 91% of patients, with a
dures (p = 0.195), best relief obtained, patient sex, patient              mean follow-up of 68 months.13 Those authors stated their
age, side of TN, previous MVD, or duration of symptoms                     final outcomes are comparable to the outcomes following
prior to first GKS procedure (p = 0.345).                                  MVD presented by Barker et al.,1 who reported freedom

180                                                                                         J Neurosurg / Volume 113 / December 2010
Repeat GKS for trigeminal neuralgia: long-term results
                                                                    TABLE 4: Trigeminal dysfunction after repeat GKS

                                                                                Complication                    No. of Cases
                                                                            trigeminal dysfunction               22 (47.8%)
                                                                            numbness                             21 (45.6%)
                                                                            bothersome numbness                   4 (8.7%)
                                                                            dry eye                               5 (10.9%)
                                                                            taste change                          4 (8.7%)
                                                                            jaw weakness                          1 (2.2%)
                                                                            anesthesia dolorosa                       0

                                                                    Huang et al. and Kondziolka et al., but we did include 2
                                                                    patients with MS, 1 patient with systemic lupus erythe-
                                                                    matosus, and 1 other patient who had acquired TN after
                                                                    an intracranial tumor resection.
                                                                         We found that the majority of treatment failures oc-
                                                                    curred by 18 months, and by 28 months, there were no
  Fig. 1. Kaplan-Meier plot showing pain relief after second GKS.   further failures. Authors of previous studies identified a
                                                                    plateau in treatment failures after repeat GKS that occurs
                                                                    at or around 18 months,3,9,10 but findings from the largest
from pain without medication at rates of 80% at 1 year              series reported by Kondziolka et al.15 showed that such
and 70% at 10 years after either 1 or 2 MVD surgeries.              failures may continue to occur for years. These results
In that series, 132 patients (11%) required second opera-           indicate that even longer follow-up is needed to fully ap-
tions. Other authors have reported less durability of pain          preciate treatment failures after second GKS.
relief after MVD. Burchiel et al.6 reported a recurrence                 A history of prior surgical treatment predicts a worse
rate after MVD of 47% in a series of 36 patients with a             outcome after GKS in many series.8,10,15,16,21,22,26 In 83% of
follow-up period averaging more than 8 years.                       cases in our study, the patient had not undergone an in-
     For patients undergoing repeat MVD, Barker et al.1             vasive surgical procedure previously. We were unable to
reported 42% excellent results (no pain and no medica-              appreciate a statistical difference between this subset for
tions) and 5% good results (≥ 75% pain relief with or               the second GKS because there were only a small number
without medications) at 10 years. Interestingly, the pres-          of patients with prior surgical intervention in our series.
ence of nonlancinating facial pain was not necessarily              Type 1 Versus Type 2 Pain
considered a criterion for failure in that series. The ef-
ficacy of a second GKS for TN in our series, as well as                  The influences of pain type on outcome after repeat
in others,3,15 appears comparable with the results after re-        GKS is not well described. In several series, atypical pain
peat MVD. One factor that may cloud a comparison with               responded less well than typical pain to initial GKS.2,15,16,26
this large MVD series is that “the presence of constant,            Moreover, Type 1 TN has been shown to be more respon-
aching, or burning facial pain that was not lancinating or          sive to MVD than Type 2 TN, perhaps because Type 1
paroxysmal was not criterion for failure.”1 In our analysis,        TN is more likely to be caused by arterial compression.17
we included all types of pain when assessing treatment              In our series, Type 1 TN did not have a significantly bet-
failure, but our requirements for success were lower. In            ter outcome than Type 2 TN after repeat GKS (p = 0.916).
short, it is difficult to compare results between studies due       There is some inherent difficulty in retrospectively com-
to variable treatment plans, pain scales, follow-up dura-           paring outcomes for Type 1 and Type 2 pain, which, as
tions, and definitions of a successful treatment. Results           described by Burchiel and colleagues, may in fact be “se-
are often published as initial response rates or recurrence         quential stages of the same disease process, with Type 2
rates rather than Kaplan-Meier survival curves.                     TN representing the clinical sequelae of more advanced
     Huang et al.13 reported a series of 20 repeat GKSs             trigeminal neuropathy.”6,17 Thus, recall bias may lead to
with a mean follow-up of 60 months. At 1, 2, 3, and 4               an overestimation of the number of patients with pretreat-
years, complete pain relief, with or without medications,           ment Type 2 pain. It is also possible that our repeat GKS
was demonstrated in 70%, 65%, 60%, and 60% of pa-                   series selected out a more treatment-resistant subset of
tients, respectively. Kondziolka et al.15 reported a series         patients with Type I pain, given that at least 1 ablative
of 72 patients with a median follow-up of 24 months af-             procedure already failed in these patients.
ter repeat GKS. They reported pain control (defined as
BNI Score IIIb = 75% reduction in pain with low-dose                Complications and Radiation Dose
medication) at 1, 3, 5, and 10 years to be 90%, 73%, 63%,               As an ablative therapy, GKS would be expected to
and 56%, respectively.15 In our series, we found adequate           have a complication profile similar to PGR, RFL, and
pain relief (≥ 50% relief) at 1, 3, 5, and 10 years to be           PBC, which produce sensory dysfunction in 50%–79%
70%, 50%, 50%, and 50%, respectively. Our series had                of cases.4,5,14,25,28 Indeed, the primary complication af-
a slightly lower response rate than that in the series of           ter GKS is new facial sensory symptoms.10,16,27 Rates of

J Neurosurg / Volume 113 / December 2010                                                                                       181
                                                                     B. Y. Kimball, J. M. Sorenson, and D. Cunningham

                                                                         relief. Using a cumulative dose of 150 Gy and our treat-
                                                                         ment plan, trigeminal dysfunction is frequent but rarely
                                                                         bothersome. Facial numbness after repeat GKS predicts
                                                                         more durable pain relief. Further study may be needed to
                                                                         better define the optimal dose for second GKSs.

                                                                                                      Disclosure
                                                                               The authors report no conflict of interest concerning the mate-
                                                                         rials or methods used in this study or the findings specified in this
                                                                         paper.
                                                                               Author contributions to the study and manuscript preparation
                                                                         include the following. Conception and design: Kimball, Sorenson.
                                                                         Acquisition of data: all authors. Analysis and interpretation of
                                                                         data: Kimball, Sorenson. Drafting the article: Kimball, Sorenson.
                                                                         Critically revising the article: Kimball, Sorenson. Reviewed final
                                                                         version of the manuscript and approved it for submission: Kimball,
                                                                         Sorenson. Statistical analysis: Kimball, Sorenson. Administrative/
                                                                         technical/material support: Cunningham. Study supervision: all
                                                                         authors.



  Fig. 2. Kaplan-Meier plot showing the relationship between numb-                                   References
ness and durable pain relief.
                                                                          1. Barker FG II, Jannetta PJ, Bissonette DJ, Larkins MV, Jho HD:
                                                                             The long-term outcome of microvascular decompression for
                                                                             trigeminal neuralgia. N Engl J Med 334:1077–1083, 1996
trigeminal dysfunction between 2% and 45% after the                       2. Brisman R: Gamma knife surgery with a dose of 75 to 76.8
initial GKS10,16,20,21,23,27,29 and between 11% and 80% af-                  Gray for trigeminal neuralgia. J Neurosurg 100:848–854,
ter a second treatment have been reported.10–13,18,19,24 We                  2004
found a 48% rate of trigeminal nerve dysfunction after 2                  3. Brisman R: Repeat gamma knife radiosurgery for trigeminal
GKS procedures with a cumulative dose of 150 Gy. There                       neuralgia. Stereotact Funct Neurosurg 81:43–49, 2003
                                                                          4. Brown JA, Gouda JJ: Percutaneous balloon compression of
were no cases of GKS-induced anesthesia dolorosa in our                      the trigeminal nerve. Neurosurg Clin N Am 8:53–62, 1997
series, but 8.7% of patients in this series reported bother-              5. Brown JA, McDaniel MD, Weaver MT: Percutaneous trigemi-
some numbness, which they believed was a good trade                          nal nerve compression for treatment of trigeminal neuralgia:
for the pain relief they attained. Pollock et al.18 reported a               results in 50 patients. Neurosurgery 32:570–573, 1993
numbness rate of 58% and a dysesthesia rate of 16% with                   6. Burchiel KJ, Clarke H, Haglund M, Loeser JD: Long-term ef-
repeat GKS using a higher cumulative dose of 163 Gy.                         ficacy of microvascular decompression in trigeminal neural-
As with the initial GKS, sensory dysfunction after repeat                    gia. J Neurosurg 69:35–38, 1988
                                                                          7. Burchiel KJ, Slavin KV: On the natural history of trigeminal
GKS predicts better pain relief.9,18,20 The results in our                   neuralgia. Neurosurgery 46:152–155, 2000
series also demonstrated a correlation between sensory                    8. Dhople AA, Adams JR, Maggio WW, Naqvi SA, Regine WF,
dysfunction and better durability of pain relief (Fig. 2).                   Kwok Y: Long-term outcomes of Gamma Knife radiosurgery
     The optimum cumulative dose with repeat GKS                             for classic trigeminal neuralgia: implications of treatment and
treatment is difficult to determine because of the small                     critical review of the literature. Clinical article. J Neurosurg
number of such cases in the literature, variability in tar-                  111:351–358, 2009
geting, and possible differences in treatment goals. In an                9. Dvorak T, Finn A, Price LL, Mignano JE, Fitzek MM, Wu JK,
effort to tailor the risks and benefits of sensory dysfunc-                  et al: Retreatment of trigeminal neuralgia with Gamma Knife
                                                                             radiosurgery: is there an appropriate cumulative dose? Clini-
tion and pain relief, Dvorak et al.9 suggested a cumulative                  cal article. J Neurosurg 111:359–364, 2009
dose lower than 130 Gy for patients with reasonable pain                 10. Hasegawa T, Kondziolka D, Spiro R, Flickinger JC, Lunsford
control while taking medications who were experienc-                         LD: Repeat radiosurgery for refractory trigeminal neuralgia.
ing intolerable medication toxicity and a higher dose for                    Neurosurgery 50:494–502, 2002
patients with debilitating refractory trigeminal pain who                11. Herman JM, Petit JH, Amin P, Kwok Y, Dutta PR, Chin LS:
may be more accepting of a higher rate of side effects.                      Repeat gamma knife radiosurgery for refractory or recurrent
The impact of shot placement with respect to the brain-                      trigeminal neuralgia: treatment outcomes and quality-of-life
                                                                             assessment. Int J Radiat Oncol Biol Phys 59:112–116, 2004
stem is still unclear. In our series, the 50% isodose line of            12. Huang CF, Chuang JC, Tu HT, Lin LY: Repeated Gamma
the first isocenter contacts the surface of the brainstem,                   Knife surgery for refractory trigeminal neuralgia. J Neuro-
whereas in other series, this first isocenter is placed more                 surg 105 (Suppl):99–102, 2006
distal so that the brainstem only receives 20%–40% of                    13. Huang CF, Tu HT, Liu WS, Chiou SY, Lin LY: Gamma Knife
the maximal dose.12,15,18                                                    surgery used as primary and repeated treatment for idiopath-
                                                                             ic trigeminal neuralgia. J Neurosurg 109  (Suppl):179–184,
                                                                             2008
                         Conclusions                                     14. Jho HD, Lunsford LD: Percutaneous retrogasserian glycerol
                                                                             rhizotomy. Current technique and results. Neurosurg Clin N 
      Repeat GKS for TN provides substantial long-term                       Am 8:63–74, 1997

182                                                                                       J Neurosurg / Volume 113 / December 2010
Repeat GKS for trigeminal neuralgia: long-term results

15. Kondziolka D, Zorro O, Lobato-Polo J, Kano H, Flannery TJ,               er BL: Gamma knife radiosurgery for trigeminal neuralgia:
    Flickinger JC, et al: Gamma Knife stereotactic radiosurgery              the initial experience of The Barrow Neurological Institute.
    for idiopathic trigeminal neuralgia. Clinical article. J Neuro-          Int J Radiat Oncol Biol Phys 47:1013–1019, 2000
    surg 112:758–765, 2010                                             24.   Shetter AG, Zabramski JM, Speiser BL: Microvascular de-
16. Maesawa S, Salame C, Flickinger JC, Pirris S, Kondziolka                 compression after gamma knife surgery for trigeminal neural-
    D, Lunsford LD: Clinical outcomes after stereotactic radio-              gia: intraoperative findings and treatment outcomes. J Neuro-
    surgery for idiopathic trigeminal neuralgia. J Neurosurg 94:             surg 102 (Suppl):259–261, 2005
    14–20, 2001                                                        25.   Taha JM, Tew JM Jr, Buncher CR: A prospective 15-year fol-
17. Miller JP, Acar F, Burchiel KJ: Classification of trigeminal             low up of 154 consecutive patients with trigeminal neuralgia
    neuralgia: clinical, therapeutic, and prognostic implications in         treated by percutaneous stereotactic radiofrequency thermal
    a series of 144 patients undergoing microvascular decompres-             rhizotomy. J Neurosurg 83:989–993, 1995
    sion. Clinical article. J Neurosurg 111:1231–1234, 2009            26.   Tawk RG, Duffy-Fronckowiak M, Scott BE, Alberico RA,
18. Pollock BE, Foote RL, Link MJ, Stafford SL, Brown PD,                    Diaz AZ, Podgorsak MB, et al: Stereotactic gamma knife sur-
    Schomberg PJ: Repeat radiosurgery for idiopathic trigeminal              gery for trigeminal neuralgia: detailed analysis of treatment
    neuralgia. Int J Radiat Oncol Biol Phys 61:192–195, 2005                 response. J Neurosurg 102:442–449, 2005
19. Pollock BE, Foote RL, Stafford SL, Link MJ, Gorman DA,             27.   Urgosik D, Vymazal J, Vladyka V, Liscák R: Gamma knife
    Schomberg PJ: Results of repeated gamma knife radiosurgery               treatment of trigeminal neuralgia: clinical and electrophysi-
    for medically unresponsive trigeminal neuralgia. J Neuro-                ological study. Stereotact Funct Neurosurg 70 (1 Suppl 1): 
    surg 93 (Suppl 3):162–164, 2000                                          200–209, 1998
20. Pollock BE, Phuong LK, Foote RL, Stafford SL, Gorman DA:           28.   Young RF: Glycerol rhizolysis for treatment of trigeminal
    High-dose trigeminal neuralgia radiosurgery associated with              neuralgia. J Neurosurg 69:39–45, 1988
    increased risk of trigeminal nerve dysfunction. Neurosur-          29.   Young RF, Vermulen S, Posewitz A: Gamma knife radiosur-
    gery 49:58–64, 2001                                                      gery for the treatment of trigeminal neuralgia. Stereotact 
21. Pollock BE, Phuong LK, Gorman DA, Foote RL, Stafford SL:                 Funct Neurosurg 70 (Suppl 1):192–199, 1998
    Stereotactic radiosurgery for idiopathic trigeminal neuralgia.
    J Neurosurg 97:347–353, 2002
22. Régis J, Metellus P, Hayashi M, Roussel P, Donnet A, Bille-          Manuscript submitted June 22, 2010.
    Turc F: Prospective controlled trial of gamma knife surgery          Accepted August 24, 2010.
    for essential trigeminal neuralgia. J Neurosurg 104:913–924,         Address correspondence to: Brent Y. Kimball, M.D., 1493
    2006                                                               Eastmoreland Avenue, Memphis, Tennessee 38104. email:
23. Rogers CL, Shetter AG, Fiedler JA, Smith KA, Han PP, Speis-        brentkimballmd@gmail.com.




J Neurosurg / Volume 113 / December 2010                                                                                            183