Microvascular decompression and percutaneous ablation surgery have historically been the treatments of choice for medical refractory trigeminal neuralgia (TGN). Gamma Knife surgery has been utilized as an alternative, minimally invasive treatment in trigeminal neuralgia.

      Our purpose was to evaluate the long-term results of Gamma surgery for the treatment of trigeminal neuralgia.

      From 1996 to 2003, we treated 151 cases of trigeminal neuralgia with Gamma Knife surgery. In this group, radiosurgery was performed once in 136 patients, twice in 14 patients, and three times in one patient. Clinical outcomes and post-radiosurgical MR imaging studies were analyzed.

      The median age of the patients was 68 (range 22-90) years old, and the median time from diagnosis to Gamma surgery was 72 months (range 1 to 276 months). The types of trigeminal neuralgia were as follows: 122 patients with typical TGN; 3 with atypical TGN; 4 with multiple sclerosis associated TGN; and 7 patients with TGN and a history of a cavernous sinus tumor. In each case, the radiosurgical target was chosen 2 to 4 mm anterior to the entry of the trigeminal nerve into the pons. The maximum doses ranged from 50 to 90 Gy.

      The median follow-up was 19 (range 2 to 96) months. The mean time to relief of pain was 24 (range 1-180) days. The percentage of patients who were pain-free without medication at 1, 2, and 3 years follow up was 47%, 45%, and 34%, respectively. The percentage of patients who experienced some degree of improvement in their pain was 90%, 77%, and 70% at 1, 2 and 3 years follow-up, respectively. Thirty-three of 122 (27%) patients with initial improvement later experienced pain recurrence at a median time of 12 months post-radiosurgery (range 2-34 months). Among those with recurrence of their symptoms, 14 patients received additional Gamma surgery, 6 had a microvascular decompression, 4 had a glycerol injection, and 1 patient had a percutaneous radiofrequency rhizotomy. Twelve (9%) patients developed the onset of new facial numbness post-radiosurgery. Post-radiosurgical MR imaging changes were noted in 9 patients (7%). In multivariate analysis, both right-sided neuralgia (p=0.032) and age (p=0.05) were statistically significant. New onset of facial numbness following Gamma surgery correlated with more than one radiosurgery (p=0.002).

      The main limitation of this study is its retrospective nature and lack of a specific control group.

      At last follow-up, Gamma Knife surgery effectuated pain relief in 44%. Some degree of pain improvement at 3 years post radiosurgery was noted in 70% of patients with trigeminal neuralgia.

      Although less effective than microvascular decompression, radiosurgery remains a reasonable treatment option for those unwilling or unable to undergo more invasive surgical approaches and offers a low risk of side effects. This data will be of value for the design of randomized trials.

      Trigeminal neuralgia (TGN) is a paroxysmal lancinating pain confined to a distribution encompassing one or more of the branches of the trigeminal nerve on one side of the face. It frequently arises in conjunction with a vascular contact between an artery or vein and the trigeminal nerve at the root entry zone (REZ). This fact led to the hypothesis of a causal relationship between vessel compression and trigeminal neuralgia and the devising of microvascular decompression surgery.¹ First line treatment for patients with TGN is medical therapy. However, many patients with this condition eventually fail medical therapy because of refractory pain or intolerable medication side-effects. More invasive treatment options include microvascular decompression (MVD), and neuronal modulating procedures such as glycerol rhizolysis, radiofrequency rhizotomy, percutaneous balloon microcompression, and peripheral nerve blocks.^1,12

      Radiosurgery has been employed to treat TGN using high-resolution imaging guidance, typically magnetic resonance imaging. Radiosurgery has been viewed as a minimally invasive treatment option with few side effects. However, questions remain about the efficacy, safety, and durability of Gamma Knife surgery for TGN.

      Our hypothesis was that through retrospective evaluation of a large clinical series , we could determine the efficacy, safety, and durability of Gamma Knife surgery for TGN.

      We evaluated our experience with Gamma Knife surgery for the treatment of 136 patients with trigeminal neuralgia and atypical facial pain and analyzed the factors associated with a pain free outcome.

      In various series, pain free outcomes have ranged from 21.8% to 75%, and complication rates varied from 2.7% to 37%.^{2,3,4,5,6,7,8,9,10, 11,13,14,15,16,17,18}

      The main limitation of our study method was that data was evaluated retrospectively using a prospective database.

      This study was approved by the University of Virginia Institutional Review Board.

      No external funding was provided for this study.

      The mean age was 68 years. From 1996 until 2004, a total of 136 patients were treated, and 122 had a diagnosis of trigeminal neuralgia while 14 had a diagnosis of atypical facial pain (Table 1).

      This patient population consisted of 66 males and 70 females.

      Ninety-four patients had right sided pain while 42 patients had left sided pain. The distribution of pain in the divisions of the trigeminal nerve was as follows: V1= 6.9% of patients; V2= 28.4%; V3= 23.3%; V1 and V2= 15.5%; V2 and V3= 15.5%; and V1 and V2 and V3= 11.2%. Pre-operative facial numbness was noted in 34 patients (25%).

      The vast majority of the patients had at some point been treated with more than one pharmacological agent, and all patients were refractory to medical treatment. In addition, seventy-four patients had undergone previous surgical procedures for treatment of their facial pain. Twenty–three (17%) underwent MVD, 46 (34%) had glycerol injections, 2 (1.5%) had radiofrequency rhizotomies, and 32 (24%) had neurectomies.

      A total of 151 Gamma Knife surgeries were performed on these 136 patients with 14 undergoing a second Gamma Knife surgery and one undergoing three Gamma surgeries. Radiosurgery was performed using the Model U (1989 to June 2001) and Model C (July 2001 to present) Gamma Knifes (Elekta Instruments, Norcross, GA). The Leksell model G head frame was applied in the main operating room under local anesthesia with light intravenous sedation (typically propofol or fentanyl and versed). Most patients were treated with one isocenter (median 1, range 1 to 7) delivered with the 4mm collimator helmet. Maximal doses ranged from 50 to 90 Gy, and the median dose was 80 Gy. The maximum prescription doses were as follows: 50 Gy in 0.7% of procedures; 60 Gy in 0.7%; 70 Gy in 43%; 80 Gy in 55%; and 90 Gy in 0.7% of cases. In each case, the root entry zone (REZ) was encompassed within the treatment plan to allow a low dose of radiation (approximately 20 to 30 Gy) to the edge of the pons. The doses less than 70 Gy were only given at retreatment. The center of the targeting was typically chosen 2 to 4 mm anterior from the junction of the trigeminal nerve and the pons. Targets were categorized into two groups. The first group of targets is when the trigeminal nerve was covered with the 50% isodose outside of the brainstem. The second category of targets was when the 50% isodose curve was adjacent to the surface of the brainstem. Fifty-two percent of patients were treated using the first target strategy and forty-three percent were treated using the second strategy.

      The trigeminal nerve was typically imaged using a 1.5 Tesla MRI unit. Localization was performed using T1-weighted and fast spin echo T2-weighted axial images along with coronal images of the nerve. The axial volume acquisition of 512x216 matrices was divided into sections of 1 mm without gaps. T1 weighted images were also repeated after administration of gadolinium. Two patients underwent computed tomography (CT) cisternography for targeting because of medical conditions (e.g. a pacemaker) that prohibited MR imaging. However, CT cisternography is not ideal for targeting purposes. 

      Clinical Follow-up of Patients 136 of 151 patients were available for follow-up. Patients received direct clinical follow-up every 3 to 6 months. If the patient failed to make his clinical appointment, a questionnaire was sent to the patient. On that questionnaire, patients were asked to assess their pain intensity, time to onset of pain relief, presence or absence of facial numbness, pharmacological agents taken for pain, and whether they underwent additional surgical interventions for treatment of facial pain. Patients were specifically asked if they were pain free or had improvements in their pain following Gamma surgery. They were also asked about the presence of new or worsened facial numbness post-operatively. Imaging Follow-up of Patients Patients received routine MRI examinations 6 to 9 months after Gamma Knife surgery. T1-weighted sequences with and without contrast and T2-weighted images were performed to evaluate for any adverse consequences. If a patient developed the onset of new facial numbness post-radiosurgery, that patient underwent additional MRI testing at the time of the onset of the numbness to evaluate for any brain tissue changes.

      All serial follow-up information was obtained via direct contact with the patient or their referring physicians.

      The tests and scoring methods were validated

      Univariate and multivariate analysis were performed to assess for variables that were predictive of a pain free outcome after Gamma Knife surgery. The following variables were assessed: right sidedness of the pain; typical or atypical pain; age; gender; number of isocenters; target distance from nerve root entry zone (either the 50% isodose well outside of or adjacent to the brainstem); maximal dose; presence of post-GK imaging changes; presence of previous interventions; a previous MVD; previous glycerol injection; previous neurectomy; increasing length of time from diagnosis to Gamma surgery; distribution of the pain; the presence of pre-operative facial numbness; the presence of post-operative facial numbness; and recurrence of pain in those patients who were treated more than once with Gamma surgery. The unpaired Student’s t-test or analysis of variance were used for continuous variables. Nominal or ordinal data was compared using the two-tailed Fisher’s exact test. All statistical analyses were conducted with the aid of Statview 5.01 (SAS Institute, Cary, NC).

      We did not use a power calculation as no control group was studied.

      P<.05 was chosen as the level of statistical significance.

      Clinical Outcome Median follow-up in the 136 patients was 19 months. The median interval from the treatment to symptom improvement was 24 (range 1-180) days. Few patients experienced a benefit more than 2 months following radiosurgery. At the last time of follow-up, 44% of patients were pain free without medication, and 56% still had some degree of pain. The percentage of patients who were pain free without medication at specific time points was as follows: 47% at 1 year (of 118 patients with 1 year follow-up); 45% at 2 years (of 64 patients with 2 years follow-up); and 34% at 3 years (of 39 patients with 3 years follow-up). In a similar fashion, those who experienced some improvement in pain post-radiosurgery were as follows: 90% at 1 year (of 118 patients with 1 year follow-up); 77% at 2 years (of 64 patients with 2 years follow-up); and 70% at 3 years (of 39 patients with 3 years follow-up). Forty-six percent of those with typical TN (n=122) were pain free at last follow-up whereas only 29% of those with atypical TN (n=14) were pain free (p=0.35, t-test). During follow-up, 33 patients had recurrence of their pain after some initial relief. The median time to recurrence of facial pain was 12 months. Among those patients with recurrence of pain, 14 patients underwent an additional Gamma Knife surgery, 6 patients had a microvascular decompression, 4 patients a glycerol injection, 1 patient a radiofrequency rhizotomy, and 5 patients refused any further treatment.

      Clinical Complications Twelve patients (9%) experienced new post-operative numbness following Gamma Knife surgery. In our study, 4 patients out of 14 who had repeat Gamma surgery (29%) had new facial numbness. Eight out of 122 patients who had only one Gamma Knife surgery (7%) developed new facial numbness. There was a statistically significant difference in the incidence of new facial numbness in these two group (p=0.002, t-test). We were unable to establish a significant relationship between radiation dose and post-operative facial numbness. No patient experienced the new onset of corneal reflex loss or anesthesia dolorosa. A total of 26 patients had either new or worsened facial numbness after Gamma surgery. The severity of the numbness was as follows: 12 patients with mild numbness; 10 with somewhat disturbing numbness; and 4 with severe numbness. Twenty-one of 122 patients (17%) receiving only one Gamma surgery developed new or worsened facial numbness while 5 out of 14 patients (36%) receiving more than one Gamma surgery experienced new or worsened symptoms. The difference was not statistically significant (p=0.14, T-test). In addition, 12 of 46 patients (26%) with a history of glycerol injection experienced new or worsened facial numbness as compared to 14 out of 90 patients (16%) without a history of glycerol injection (p=0.14, T-test). One patient described that the facial numbness was worse than the original TGN pain because it led to drooling. She did not have anesthesia dolorosa. Imaging Outcome Nine patients (6.6%) had post-radiosurgical imaging changes including focal contrast enhancement or focal T2-weighted signal change. Among those nine patients who had post-operative imaging changes, 4 had no numbness, 3 had mild numbness, and 2 had severe numbness.

      In order to evaluate the efficacy of Gamma radiosurgery in a rigorous fashion, we chose to consider factors that were statistically related to a pain free outcome at the last time of follow-up. Seventeen factors in all were tested. In univariate analysis, right sidedness to the pain (p=0.0002) and a previous neurectomy (p=0.04) were statistically related to a pain free result. In multivariate analysis, again right sidedness (p=0.032), and this time increasing age (p=0.05) were related to a pain free result. No other factors were significantly related (i.e. p>0.05) to a pain free result following Gamma Knife surgery.

             The underlying etiology of trigeminal neuralgia has been the subject of intense investigation for more than a century. Research by Pfluger (1859) and Heidenhain (1861) demonstrated that the proximal end of a nerve was more excitable than the distal end and that the excitability of the nerve increased near the region of a cross section. By 1941, Olivecrona described that mechanical pressure along the root or at the level of the ganglion could be the cause of trigeminal neuralgia. Granit, Leksell, and Skoglund (1944) demonstrated that local pressure on nerve fibers could result in painful afferent discharges from the injured neural segment. Jannetta and others have suggested that vascular compression of the trigeminal nerve may be a causal agent in trigeminal neuralgia.^1,12 Following the successful long-term outcome of two patients who had trigeminal neuralgia radiosurgery using ortholvoltage x-rays Leksell and Hakanson treated 48 patients between 1970 and 1978, and follow-up information was available on 46 of these patients . In the first 24 patients, plain stereotactic skull X-rays were used for targeting, and 33% were pain free at 6 months while only 8.3% were pain free at a mean follow up of 26 months (Personal communication, Sten Hakanson, 2004). In the second set of 22 patients, the position of the Gasserian ganglion was determined by transoval cisternography using tantalum dust. For that group, the percentages of patients who were pain free at 6 and 26 months were 59% and 18%, respectively. Overall, the treatment of choice for trigeminal neuralgia remains microvascular decompression. However, for patients who are ineligible for or unwilling to undergo more invasive neurosurgical procedures, Gamma Knife surgery offers a reasonable alternative. Gamma Knife surgery also does not carry the same type and degree of risks that microvascular decompression does. For instance, in a long-term series of microvascular decompression patients, there were the following risks from the series by Dr. Jannetta: 0.2% death; 0.1% brain stem infarct; and 1% hearing loss.¹ None of these complications were observed in our series nor were they associated with radiosurgical treatment of trigeminal neuralgia in other major centers. Ultimately, the patient must choose the type of intervention he or she is willing to undergo.

      Measuring Pain Relief We believe it is better to grade patient outcome in a straightforward fashion by reporting pain free outcomes off medications. Such an endpoint is readily understandable to patients, neurosurgeons, and referring physicians of all backgrounds. Reporting outcomes in terms of subjective degrees of pain relief may lead to confusion for patients and physicians alike. As a means of comparing our results to other studies which report pain relief rather than pain free outcomes, we do note the numbers of patients who reported improvements in their trigeminal neuralgia. In our study, pain relief meant that the patients were still on medications but thought their pain was improved after Gamma Knife surgery. This result is arguably very meaningful to patients with severe pain and can translate into a significant improvement in the quality of life. The variable history of trigeminal neuralgia is the other major difficulty in analyzing results from small to medium size retrospective studies. Trigeminal neuralgia can be characterized by spontaneous partial or complete remissions. The characteristic waxing and waning nature as well as the subjectivity of pain, sensory loss, and paresthesias makes a longer follow-up period and straightforward endpoints essential. Prognostic Factors for Pain Relief In our study, multivariate analysis revealed that right sided pain and age correlated with a pain free outcome. In our study, it is notable that a pain free outcome was not related to dose, sensory loss, slight differences in target selection, or the type of pain (i.e. either atypical or typical trigeminal neuralgia).
The relationship between a previous neurectomy and Gamma Knife surgery in terms of a pain free outcome suggests the efficacy of the Gamma Knife may not be reduced after a peripheral neurectomy. Clearly, the Gamma Knife and a peripheral neurectomy target proximal and distal portions of the symptomatic trigeminal nerve, respectively. In a subset of patients, it is possible that both proximal and distal portions of the nerve should be lesioned to provide adequate pain relief. The effect of age on outcome may in part be due to the generally older age of the patient treated with Gamma Knife (median age of 68) and the fact that many of these patients had had Gamma surgery as a first line treatment. Other studies have reported different factors that were associated with a better response to Gamma Knife surgery. These include typical trigeminal neuralgia rather than atypical pain from multiple sclerosis or other causes¹³, higher doses of radiation³, a target closer to the brainstem, and no prior surgery⁶. Another favorable prognostic factor in patients with no prior surgical intervention is MRI evidence of blood vessel contact with the trigeminal nerve.⁴ In the present study, we did not find these factors related to a pain free outcome.

      Complications of Gamma Knife surgery The most frequent complication following radiosurgical treatment of trigeminal neuralgia is facial numbness. The incidence of new trigeminal dysfunction varies from 6% to 66%.^{2,3,4,5,8,14,15} In this series, 12 out of 136 patients (9%) developed new facial numbness following Gamma Knife surgery. In our study, only 1 patient received a dose of 90 Gy and no facial numbness was noted in this case. Fortunately, we did not observe cases of anesthesia dolorosa or absence of the corneal reflex in the 136 patients. In our study, 29% of patients who had repeat Gamma Knife surgery had new facial numbness whereas 7% of patients who had only one Gamma Knife surgery developed new facial numbness (p=0.002, t-test). In a series of 18 patients who underwent repeat radiosurgery, Herman et al. (2004) noted an 11% incidence of new or worsened facial numbness which was not substantially elevated over the risk of facial numbness for those having only one Gamma Knife surgery.⁷ However, Hasegawa et al. (2002) noted an increased risk of facial numbness associated with repeat Gamma surgery.⁶

      Gamma Knife surgery is a relatively safe and effective treatment option for patients with medically refractory trigeminal neuralgia. The improvements in pain following Gamma surgery diminish somewhat with time, but this is true for other treatments as well. Studies of long term durability are important. Future work should consider randomized trials between the different therapeutic options in patients appropriate for randomization.