Radiosurgery for Trigeminal Neuralgia
Douglas Kondziolka, MD
Professor of Neurological Surgery and Radiation Oncology
Suite B-400, UPMC Presbyterian, 200 Lothrop Street, Pittsburgh, PA 15213
Telephone: 412-647-6782, Fax: 412-647-0989, firstname.lastname@example.org
Medical therapy is the initial approach for most patients with trigeminal neuralgia. However, many patients fail or cannot tolerate medical therapy, and eventually require surgical intervention. Although often associated with initial pain relief, all surgical procedures have variable but definite rates of recurrence and morbidity. Stereotactic radiosurgery (Gamma knife® radiosurgery, linear accelerator radiosurgery) has been advocated as a minimally invasive alternative surgical approach to microvascular decompression or percutaneous surgeries (15). The majority of data has been published using Gamma knife technique. Stereotactic irradiation of the trigeminal ganglion was first reported by Leksell (17). More recently, numerous authors have reported their initial results with radiosurgery using high-resolution image guidance (11,13,14,18,20,22-24,26,29,31). This data testified to the accuracy of radiosurgical targeting, the initial safety of the technique, and provided dose-response information.
Technique of Radiosurgery
Typically, patients undergo the entire procedure during a short outpatient hospital stay and are discharged home the same day. First, the Leksell model G stereotactic frame (Elekta Instruments, Atlanta GA) is attached to the head under local anesthesia. Second, the patient undergoes stereotactic magnetic resonance imaging (MRI) to identify the trigeminal nerve. Rare patients have computed tomography (CT) imaging for targeting if they cannot undergo MRI. MRI is performed using contrast-enhanced, short repetition time (TR) sequences and axial volume acquisitions (512 x 216 matrix) divided into 1mm slices. When the trigeminal nerve is difficult to identify on imaging (usually because of prior microvascular decompression), additional axial long relaxation time MR images were used (9). At present, a single 4-mm isocenter is used for targeting. With a single isocenter, the target is 3 to 6 mm anterior from the junction of the trigeminal nerve and pons (Fig. 1). The isocenter is usually located so that brainstem edge is irradiated at the 20 or 30% isodose line. Past patients received maximum doses varying from 60 to 90 Gy, although 80 Gy is a common maximal dose.
Method of Patient Evaluation
After radiosurgery, we evaluate patients for the degree of pain relief, latency interval to pain relief, need for further surgical procedures, use of medication, and complications. In addition, up-to-date clinical information is obtained by telephone interviews at periodic intervals (20). To evaluate effectiveness over time (besides the typical measure of the proportion of patients with pain relief at initial or last follow up), we calculate the time to initial response from radiosurgery and the duration of pain relief using the product limit method of Kaplan and Meier. The time to onset of complications after radiosurgery (new or increased sensory disturbance) is also calculated using the same methods. Pain relief duration is calculated from the time when the level of pain relief being studied (complete or >50%) was achieved. We correlate outcomes (time to initial response, duration of benefit, and complications) with treatment parameters that include sex, age, duration of symptoms, presence or absence of preoperative paresthesia or additional atypical features, number of prior surgeries, maximum dose, and number of isocenters.
The University of Pittsburgh Experience
Our current experience includes more than 750 patients, managed since 1992. There were 60% women. The mean age was 68 years (range, 16-92). The mean duration of symptoms was 8 years.
Our last detailed review studied 264 consecutive radiosurgery procedures for trigeminal neuralgia, all performed between 1992 and 1998 (20). Of these 264 procedures, 25 were performed for non – idiopathic trigeminal neuralgia (19 were associated with multiple sclerosis (MS), three were secondary to tumors, and three were secondary to other pathological conditions such as a herpetic infection, Lyme disease or Wallenberg’s syndrome).
These 25 procedures were excluded from this study. We performed 239 procedures for idiopathic trigeminal neuralgia; 19 procedures represented repeat radiosurgery after an initially successful procedure. For these 19 patients, the outcome after initial radiosurgery was included in this study as a poor result (subsequent outcomes of the repeat procedure were excluded). Thus, 220 procedures in 220 patients were available for analysis. All 220 patients had trigeminal neuralgia that was idiopathic, long standing, and refractory to medication therapy such as carbamazepine, phenytoin, baclofen, or gabapentin as well as a variety of analgesic medications.
In these 220 patients, 94 were male and 126 were female. The median patient age was 70 years (range: 26-92). Most of the patients had a long history of medical treatment with the median symptom duration of 96 months (range; 3-564 months). Pain was predominantly distributed in the V2 and V3 distributions of the trigeminal nerve (29.5%), followed by V2 alone (22.3%) and V3 alone (13.2%). Patients with first division pain often had radiosurgery as an alternative to other procedures. Although all patients complained of the typical trigeminal neuralgia features of a sharp, lancinating or shooting electric-shock type of pain triggered by a variety of events, sixteen (7.3%) complained of additional features such as a more constant dull, aching or burning pain. Eighty patients (36.4%) had some sensory disturbance (usually paresthesias) preoperatively, and three patients (1.4%) had partial deafferentation pain caused by prior ablative procedures.
Prior surgery was performed in 135 patients (61.4%), including microvascular decompression, glycerol rhizotomy, radiofrequency rhizotomy, balloon microcompression, peripheral neurectomy, or ethanol injections. In these 135 patients, 86 (39.1%) had one, 39 (17.7%) had two, and 10 (4.5%) had three or more procedures prior to radiosurgery. Thus, the majority of patients represented both medical and surgical failures. In the remaining 85 patients (38.6%), radiosurgery was the first surgical procedure performed.
The outcome of the intervention was graded into 4 categories: excellent, good, fair and poor. Complete pain relief without the use of any medication was defined as an excellent outcome. We recommended all patients with complete pain relief to taper off their medications, and some patients were in the process of tapering at the time of evaluation (or refused to taper off because of the fear of a recurrence). Those patients with complete pain relief but who were still using some medication were considered as good outcomes. Patients with partial pain relief (more than 50% pain relief) were considered to have a fair outcome (11). No pain relief or less than 50% pain relief were considered as ‘’ poor’’. Placement within a category was decided by the patient rather than by the physician. Criteria for improvement included a reduction in both the frequency and severity of pain attacks. Of the 220 patients, 47 (25.1%) required further additional surgical procedures because of poor pain control. These patients were considered as treatment failures (poor outcome), and the results after the additional procedure were excluded from this analysis.
Most of the patients responded to radiosurgery within 6 months of the procedure (median, 2 months). The first evaluation was performed for all patients within 6 months after radiosurgery. At the initial follow-up assessment, excellent results were obtained in 105 patients (47.7%), and excellent plus good results were found in 139 patients (63.2 %). More than 50% pain relief (excellent, good, or fair) was noted in 181 patients (82.3%). At the last follow-up evaluation, 88 patients (40%) had excellent outcomes, 121 patients (55.9%) had excellent plus good outcomes, and 152 were fair or better (69.1%). Thirty patients (13.6%) had recurrence of pain after the initial achievement of pain relief (25 patients after complete relief, 5 patients after more than 50% relief) between 2 and 58 months after radiosurgery. Recurrences occurred at a mean of 15.4 months from irradiation.
Time to the Initial Response
The time to response after radiosurgery was analyzed using the product limit method of Kaplan and Meier. The median time to achieving more than 50% pain relief (excellent, good or fair) was 2 months (2.0 +0.05), and median time to achieving complete pain relief (good or excellent) was also 2 months (2.0+5.1). At 6 months after treatment, 81.4+2.6% patients had achieved more than 50% pain relief, and by 12 months, 85.6+2.47% (actuarial statistics). Complete pain relief (good or excellent) was achieved in 64.9+3.2% of the patients at 6 months, 70.3+3.16% by one year, and in 75.4+3.49% of patients by 33 months.
Prior authors, including our group, noted a latency interval to pain relief of approximately one to two months (11,26,31). However, approximately 15% of patients had no improvement in their pain even after 12 months. Because no patient achieved complete or even more than 50% pain relief after 12 months follow-up, we believe that patients with unchanged pain at one year cannot expect any improvement afterwards. In reality, this is an uncommon situation since most patients would not wait so long to attain pain relief. Patients who have continued disabling pain despite radiosurgery are managed with other surgical procedures. We advocate repeat radiosurgery only if complete pain relief had been achieved with subsequent recurrence.
Complete pain relief was achieved at a median time of two months with most patients achieving this level of relief within 6 months. Interestingly, a further 10% of patients achieved complete pain relief 6 to 33 months after radiosurgery. All of these patients obtained partial pain relief within 6 months, then had complete resolution of pain afterwards. Thus, it appears that patients with partial pain relief may go on to complete relief over time. This finding provides some ideas regarding the mechanism of response to radiosurgery. Since most patients responded in the initial months, electrophysiologic blocking of ephaptic transmission after nerve irradiation, which has been discussed as a possible mechanism by some authors (14,31), is possible. On the other hand, since some patients responded or improved in a later phase (more than 6 months after radiosurgery), a later radiation effect with axonal degeneration may be present.
Maintenance of Pain Relief
The duration of pain relief after initial response in all patients was also analyzed. Patients who never responded to radiosurgery were recorded as having a relief duration of zero months. More than 50% pain relief (excellent, good or fair) was achieved and maintained in 75.8+2.9% of patients at 1 year, 71.3+3.3% of patients at 2 years, 67.2+3.9% of patients at 3 years, 65.1+4.3% of patients at 3.5 years, and 55.8%+9.3% of patients at 5 years. Complete pain relief (excellent or good) was achieved and maintained in 63.6+3.3% of patients at 1 year, 59.2+3.5% of patients at 2 years, and 56.6+3.8% of patients at 3 years. A history of no prior surgery was the only factor significantly associated (p = 0.01) with achieving and maintaining complete pain relief.
Side Effects of Radiosurgery
No patient sustained an early complication after any radiosurgery procedure. Seventeen patients (7.7%) developed increased facial paresthesiae and/or facial numbness that lasted more than 6 months. Others have noted a dry eye, without significant facial numbness (21). The median time to developing paresthesiae was 8 months (range, 1 - 19 months). After 19 months, no patient developed any new sensory symptoms. No patient developed a mastication deficit after radiosurgery or noted problems in facial motor function. One patient (0.4%) developed deafferentation pain after radiosurgery. This patient had recurrent trigeminal neuralgia previously treated by microvascular decompression (MVD). Following her MVD she had some decrease in facial sensation, and she had the additional atypical feature of constant burning. Eight years after initial surgery, radiosurgery was performed to a maximum dose of 75 Gy using two isocenters. Eight months later, she developed increased constant burning pain and sensory loss consistent with deafferentation pain.
Comparing Radiosurgery to Other Procedures
The results of radiosurgery do not appear to be as good as those observed after a first microvascular decompression, but appear to be as good or better as a second procedure. Barker et al, in reporting on Jannetta’s series of 1185 patients, found that complete pain relief was maintained in 70% of patients at 10 years (1). Thus, we continue to advocate microvascular decompression for appropriate patients with trigeminal neuralgia because of the potential for longer duration pain relief (19). Nevertheless, the possible risks of MVD make it unsuitable for some patients with advanced age or medical comorbidities. In addition, the benefit of MVD is less when performed a second time, or for recurrent trigeminal neuralgia. Imaging studies after MVD, particularly if the patient develops sensory loss, frequently show small infarcts within the lateral pons (9). We consider radiosurgery a good choice for patients with recurrent pain after MVD or percutaneous surgery has failed, even though prior surgical failure reduces the radiosurgical success rate.
Other investigators have reported long term results following different percutaneous techniques for trigeminal neuralgia. Percutaneous retrogasserian thermal rhizotomy was first described in 1974 by Sweet and Wepsic and has been performed widely (27). In Broggi’s series of 1000 patients with 9.3 years of mean follow-up, they found an initial pain relief rate of 95%, with recurrence in 18.1% (2). Scrivani et al reported their series of 215 patients and found that 83% maintained pain relief at a mean interval of 32 months (25). Percutaneous retrogasserian glycerol rhizotomy, first reported by Håkanson, is also widely used (7). Jho and Lunsford reported that 90% of 523 patients achieved complete pain relief initially and that 77% maintained longer term pain control, sometimes requiring multiple procedures (10). The pain recurrence rate was estimated to be between 30% and 50% over 2 to 10 years. Others reported a longer-term pain control rate of 50-90% (6,7). Results following percutaneous trigeminal nerve balloon compression indicated a high rate of initial pain relief but that pain recurred in 26% (50 patients with average 3-year follow-up)(3,4).
The low incidence of complications is the greatest advantage of stereotactic radiosurgery compared to all other surgical options. Paresthesia or numbness of varying magnitude is observed in 20% to 70% of patients after percutaneous thermorhizotomy, glycerol rhizotomy or balloon nerve compression (2,3,6,7,10,19,25,27). In this study, less than 10% of patients developed increased facial paresthesias and/or facial sensory loss. The majority of our patients described their numbness or paresthesias as minor and not bothersome. Some authors advocated percutaneous balloon nerve compression as advantageous for management of first division trigeminal neuralgia, because of the lower risk for postoperative corneal analgesia. In the present radiosurgery study, no patient developed this complication. The reduced rate of facial sensory deficits or symptoms indicates that the radiosurgical effects on nerve tissue may be less than after other ablative surgeries. Thus, the mechanistic effect of radiosurgery is likely a combination of both histologic and electrophysiologic responses.
Radiosurgery can be repeated if pain returns after initial relief. We advocate a maximum dose of 50-60 Gy at a second procedure, and usually target a volume anterior to the prior target. Doing so has led to a pain response similar to that after primary radiosurgery in properly selected patients (8).
How does Radiosurgery Work?
In order to study the effects of radiosurgery on trigeminal nerve fibers, we used a baboon radiosurgery model and delivered either 80 or 100 Gy to the nerve using a single 4mm isocenter (12). Trigeminal nerves that received 80 Gy radiosurgery had no inflammation. Focal myelin pallor and vacuolation without fibrosis was noted. Immunoreactivity for neurofilament revealed substantial axonal loss, fragmentation, and some axonal swellings. Plastic sections and ultrastructural studies revealed a focal region near the center of the nerve exhibiting acutely degenerating axons with preservation of some identifiable myelinated axons admixed with small foci of tissue necrosis. The surrounding region contained only rare degenerating axons in which the trigeminal ganglia appeared normal as did the distal nerve beyond the radiosurgery target.
Through partial (focal) axonal degeneration, radiosurgery likely relieves trigeminal neuralgia pain by impacting upon an axon population large enough to result in pain relief. On the other hand, the low incidence of lost facial sensation indicates that the remaining intact axonal population is enough to maintain neurologic function in the majority of patients. This balance between pain relief and preservation of sensation is dose-related.
Improving Radisurgery Technique
We compared radiosurgery using two 4mm isocenters to that using one isocenter with a maximum dose of 75 Gy. Our hypothesis was that irradiation of a longer nerve segment may have led to an improved rate of pain relief. This randomized, blinded trial found no pain benefit to two isocenter radiosurgery, but a higher rate of facial sensory loss (5). Regis et al advocated a maximum dose of 90 Gy and a more anterior radiosurgery target (24). Doing so, they have reported a low rate of facial sensory loss despite using a higher dose. Reports from Young et al and Pollock et al found a higher rate of sensory loss at 90 Gy, although they used the more traditional posterior nerve target (23). These questions, dose and target, are appropriate considerations for randomized clinical trials. There are numerous papers published now from centers such as the University of Southern California, University of Virginia, State University of New York (Buffalo), and Erasmus University in Brussels. Proper patient selection is the key, with accurate radiosurgical technique. Most centers use a maximum dose of 80 to 90 Gy targeted with a single 4mm isocenter.
- Barker FG, Jannetta PJ, Bissonette DJ, et al: The long - term outcome of microvascular decompression for trigeminal neuralgia. New Engl J Med 334: 1077 – 1083,1996.
- Broggi G, Flanzini A: Long - term results of percutaneous retrogasserian thermorhizotomy for “essential” trigeminal neuralgia; Considerations in 1000 consecutive patients. Neurosurgery 26: 783 – 787, 1990.
- Brown JA, McDaniel MD, Weaver MT: Percutaneous trigeminal nerve compression for treatment of trigeminal neuralgia; Results in 50 patients. Neurosurgery 32: 570 – 573, 1993.
- Brown JA, Hoeflinger B, Long PB, et al: Axon and ganglion cell injury in rabbits after percutaneous trigeminal balloon compression. Neurosurgery 38: 993 – 1004,1996.
- Flickinger JC, Pollock BE, Kondziolka D, Phuang LK, Foote R, Stafford S, Lunsford LD. Does increased nerve length within the treatment volume improve trigeminal neuralgia radiosurgery? A prospective, blinded, randomized study. Int J Radiat Oncol Biol Phys 51:449-454, 2001.
- Fujimaki T, Fukushima T, Miyazaki S: Percutaneous retrogasserian glycerol injection in the management of trigeminal neuralgia. Long - term follow - up results. J Neurosurg 73: 212 – 216, 1990.
- Hakanson S: Trigeminal neuralgia treated by injection of glycerol into trigeminal cistern. Neurosurgery 9: 638 – 646, 1981.
- Hasegawa T, Spiro R, Flickinger JC, Kondziolka D, Lunsford LD. Repeat radiosurgery for trigeminal neuralgia. Neurosurgery 50:494-502, 2002.
- Jawahar A, Kondziolka D, Kanal E, Bissonette D, Lunsford LD. Imaging the trigeminal nerve and pons before and after surgical intervention for trigeminal neuralgia. Neurosurgery 48:101-107, 2001.
- Jho HD, Lunsford LD: Percutaneous retrogasserian glycerol rhizotomy; Current techniques and results. Neurosurg Clinics N Amer 8 (1): 63 – 74, 1997.
- Kondziolka D, Lunsford L, Flickinger JC, et al: Stereotactic radiosurgery for trigeminal neuralgia. A multiinstitutional study using gamma knife unit. J Neurosurg 84: 940 – 945, 1996.
- Kondziolka D, Lacomis D, Niranjan A, Maesawa S, Mori Y, Fellows W, Lunsford LD. Histologic effects of trigeminal nerve radiosurgery in a primate model: Implications for trigeminal neuralgia radiosurgery. Neurosurgery 46:971-977, 2000.
- Kondziolka D, Lunsford L, Flickinger JC: Gamma knife radiosurgery as the first surgery for trigeminal neuralgia. Stereotact Funct Neurosurg 70 (suppl 1): 187 – 191, 1998.
- Kondziolka D, Lunsford LD, Flickinger JC. Stereotactic radiosurgery for the treatment of trigeminal neuralgia. Clin J Pain 18:42:47, 2002.
- Kondziolka D: Functional radiosurgery. Neurosurgery 44: 12 – 21, 1999.
- Kondziolka D, Lunsford LD, Flickinger JC: The radiobiology of radiosurgery. Neurosurg Clin N Amer 10 (2): 157 – 166, 1999.
- Leksell L: Stereotaxic radiosurgery in trigeminal neuralgia. Acta Chir Scand 37: 311 – 314, 1971.
- Lindquist C, Kihlstrom L, Hellstrand E: Functional neurosurgery - A future for the gamma knife? Stereotact Funct Neurosurg 57: 72 – 81,1991.
- Lunsford LD, Apfelbaum RI: Choice of surgical therapeutic modalities for treatment of trigeminal neuralgia; Microvascular decompression, percutaneous retrogasserian thermal, or glycerol rhizotomy. Clin Neurosurg 32:319 – 333, 1985.
- Maesawa S, Salame C, Pirris S, Flickinger JC, Kondziolka D, Lunsford LD. Clinical outcomes after stereotactic radiosurgery for idiopathic trigeminal neuralgia. J Neurosurg 94:14-20, 2001.
- Matsuda S, Serizawa T, Sato M, Ono J. Gamma knife radiosurgery for trigeminal neuralgia: the dry-eye complication. J Neurosurg (Suppl 5)97:525-528, 2002.
- Pollock BE, Phuong L, Gorman D, Foote R, Stafford S. Stereotactic radiosurgery for idiopathic trigeminal neuralgia. J Neurosurg 97: 347-353, 2002.
- Pollock B, Phuong L, Foote R, Stafford S, Gorman D. High-dose trigeminal neuralgia radiosurgery associated with increased risk of trigeminal nerve dysfunction. Neurosurgery 49:58-64, 2001.
- Regis J, Metellus P, Dufour H, Roche P, Muracciole X, Pellet W, Grisoli F, Pergaut JC. Long-term outcome after gamma knife surgery for secondary trigeminal neuralgia. J Neurosurg 95: 199-205, 2001.
- Scrivani JS, Keith DA, Mathews ES: Percutaneous stereotactic differential radiofrequency thermal rhizotomy for the treatment of trigeminal neuralgia. J Oral Maxillofac Surg 57: 104 – 111, 1999.
- Shetter A, Rogers C, Ponce F, Fiedler J, Smith K, Speiser B. Gamma knife radiosurgery for recurrent trigeminal neuralgia. J Neurosurg (Suppl 5)97:536-538, 2002.
- Sweet WH, Wepsic JG: Controlled thermocoagulation of trigeminal ganglion and rootlets for differential destruction of pain fibers. J Neurosurg 40: 143 –156, 1974.
- Thompson TP, Maitz AH, Kondziolka D, et al: Radiation, radiobiology, and neurosurgery. Contemp Neurosurg 21 (13): 1 – 6, 1999.
- Urgosik D, Vymazal J, Vladyka V, et al: Gamma knife treatment of trigeminal neuralgia; Clinical and electrophysiological study. Stereotact Funct Neurosurg 70 (suppl 1): 200 – 209, 1998.
- Van der Kogel AJ: Radiation - induced damage in the central nerve system; An interpretation of target cell responses. Br. J. Cancer 53 (suppl. 7): 207 – 217, 1986.
- . Young RF, Vermeulen SS, Grimm P, et al: Gamma knife radiosurgery for treatment of trigeminal neuralgia; Idiopathic and tumor related. Neurology 48: 608 – 614, 1996.