ObjectRepeated Gamma Knife surgery (GKS) for trigeminal neuralgia (TN) is an acceptable method for refractory cases but not well established in terms of dose effect and nerve tolerance. The authors report their experience in 28 patients over 3.5 years of follow up.MethodsBetween 1999 and 2004, a total of 28 patients with recurrent or residual TN underwent repeated GKS. The median follow-up periods were 52 and 43 months after the first and repeated procedures, respectively. The entry zone of the trigeminal nerve was targeted using a 4-mm collimator and treated with 40 to 76 Gy as maximal dose. Additive doses ranged between 110 and 152 Gy. The median duration of symptoms was 4.86 years. There were 12 women (46%) and 16 men (54%).At the last evaluation, a total of 19 patients (68%) reported pain relief. Of these patients, 13 were no longer taking pain medications. Significant recurrent or residual pain was noted in nine patients after a median follow up of 12 months (range 6–48 months). New onset of facial numbness was noted in 10 patients. An additive dose above 115 Gy was found to be associated with facial numbness and nonfacial numbness (p = 0.047). No definite additive dose correlation with pain relief was noted (p = 0.23).Conclusions Repeated GKS established durable pain relief in a majority of patients, and a higher additive dose (> 115 Gy) tended to cause facial numbness. However, a prospective trial is needed to fully assess the efficacy and late complications of GKS.

ObjectThe results of microsurgery in patients with trigeminal neuralgia (TN) after failure of repeated Gamma Knife surgery (GKS) have not previously been examined. The effects of microvascular decompression (MVD) in patients in whom repeated GKS has failed are reported.MethodsBetween 1999 and 2004, eight patients (five women and three men) who underwent repeated GKS and experienced treatment failure subsequently underwent MVD. These patients underwent MVD for recurrent or residual intractable trigeminal pain. The duration of facial pain before MVD ranged from 0.8 to 13 years (mean 5.2 years). The total treatment dose administered ranged from 110 to 140 Gy (mean 129 Gy). The mean age was 59.5 years (range 44–68 years). Seven patients in whom the offending vessel (five arteries and two veins) was located at the nerve entry zone improved with variable pain relief after MVD. One patient in whom the vessel could not be identified did not improve. Intraoperatively, there were no signs in any patient of radiation-induced damage such as thickening of arachnoid membrane. In one patient a segment of atherosclerotic plaque was discovered at the nerve entry zone. Two patients experienced numbness at 1 and 3 months after MVD.Conclusions Microvascular decompression is an effective treatment for patients with TN in whom repeated GKS has failed. Previous radiation treatment did not negatively impact operative difficulty or risk to the patient in this small series. Radiosurgery may induce vascular injury and cause pain recurrence.

Object Cellular density is a major factor for change in the apparent diffusion coefficient (ADC). The authors hypothesized that loss of tumor cells after Gamma Knife surgery (GKS) may alter the ADC value and used diffusion weighted MR imaging (DW imaging) to evaluate cellular changes in brain tumors to detect their treatment response and the efficacy of GKS. Methods In this paper the authors describe a prospective trial involving 86 patients harboring 38 solid or predominantly solid brain metastases, 30 meningiomas, and 24 acoustic neuromas that were treated by GKS. The patients underwent serial MR imaging examinations, including DW imaging, before treatment and at multiple intervals following GKS. Follow-up MR images and clinical outcomes were reviewed at 3-month intervals for metastatic lesions and at 6-month intervals for benign tumors. Apparent diffusion coefficients were calculated from echoplanar DW images, and mean ADC values were compared at each follow-up. Results The mean ADC value for all meningiomas was 0.82 ± 0.15 × 10−3 mm2/sec before GKS. The mean ADC value as of the last mean follow-up of 42 months was 1.36 ± 0.19 × 10−3 mm2/sec, a significant increase compared to that before treatment (p < 0.0001). Calcification (p = 0.006) and tumor recurrence (p = 0.025) significantly prevented a rise in the ADC level. The mean ADC value for all solid acoustic neuromas was 1.06 ± 0.17 × 10−3 mm2/sec before GKS. The mean ADC value as of the last mean follow-up of 36 months was 1.72 ± 0.26 × 10−3 mm2/sec, a significant increase (p = 0.0002) compared with values before GKS. At the last mean MR imaging follow-up there appeared to be tumor enlargement in 3 patients (12.5%); however, since the ADC values in these patients were significantly higher than the preradiosurgery values, the finding was considered to be a sign of radiation necrosis rather than tumor recurrence. The mean ADC value of metastatic tumors was 1.05 ± 0.12 × 10−3 mm2/sec before GKS. This value rose significantly (p < 0.0001) to 1.64 ± 0.18 × 10−3 mm2/sec after GKS at a mean follow-up of 9.4 months. Magnetic resonance imaging showed that 89% of these tumors had been controlled by GKS. In 2 patients there were enlarged lesions, but the ADC values were the same as pre-GKS levels, and therefore, the lesions were deemed recurrent. Conclusions Apparent diffusion coefficient values may be useful in evaluating treatment results before a definitive change in volume is evident on imaging studies. In some patients in whom imaging findings are equivocal, ADC values may also be used to distinguish radiation-induced necrosis from tumor recurrence.