Brain stem metastases treated with radiosurgery: prognostic factors of survival and life expectancy estimation.

BACKGROUND: The objective of this study is to study prognostic factors of survival and 3 stratification systems for life expectancy estimation in patients with brain stem metastases treated with radiosurgery. METHODS: Between December 1999 and November 2006, 25 patients with 27 brain stem metastases were treated with Gamma Knife radiosurgery. The lesions' mean volume was 0.6 mL (0.013-3.6 mL). The mean marginal dose was 20 Gy (15-24 Gy). Univariate and multivariate studies were done to identify prognostic factors, and 3 patient stratification systems were applied for survival estimation: RPA, SIR, and BSBM. RESULTS: The primary tumor location was in the lungs in 12 patients, breast in 8, and other in 5. Fourteen lesions were located in the pons, 9 in the midbrain, and 4 in the medulla. All patients were followed clinically. Radiologic follow-up was available in 21 lesions (78%). Tumor control was achieved in all but one followed lesion (95%). There were no complications related to treatment. Median survival of patients with brain stem metastases was 11.1 months. In multivariate analysis, KPS of 80 or more, control of the primary tumor, absence of radiotherapy, and a marginal dose higher than 18 Gy were associated with better survival. The BSBM in the univariate and multivariate analyses was the strongest predictor of survival (P < .0001). CONCLUSIONS: The BSBM was the most useful tool for estimating survival. Rather than the brain stem location of an intracranial metastasis, the patient integral clinical status seems to be more important in determining survival.

Neurovascular compression anatomy and pain outcome in patients with classic trigeminal neuralgia treated by radiosurgery.

OBJECTIVE: To study the influence of the anatomy of neurovascular compression (NVC) on pain outcome in patients with classic trigeminal neuralgia treated by radiosurgery. METHODS: Analysis of the anatomy of the trigeminal nerve, brainstem, and vessels was performed in 89 consecutive patients treated by Leksell gamma knife (Elekta Instruments, Stockholm, Sweden) for classic trigeminal neuralgia. One-millimeter axial magnetic resonance imaging slices (T1-weighted, T1-weighted enhanced, and T2-weighted selected partial inversion recovery) with coronal, sagittal, and three-dimensional reconstructions were viewed. The end point for outcome was total pain remission and no medication. The follow-up period ranged between 6 and 42 months. RESULTS: In 82 patients (92%), a vascular structure in contact with the nerve was observed. In four patients, the NVC was produced by a large vessel (basilar or vertebral artery) and in 78 by a smaller vessel. The superior cerebellar artery was the cause of the NVC in 64 patients (78%). The NVC was proximal (<3 mm to the brainstem) in 34 patients (41%) and distal in 43 patients (52%). Nerve dislocation was present in 29 patients (33%), and nerve atrophy was present in 21 (24%) patients. Visualization of NVC on magnetic resonance imaging scans was not associated with outcome. The two variables associated with poor outcome were a large vessel contacting the nerve with brainstem deformation and proximal NVC. Nerve atrophy and nerve dislocation were not associated with outcome. CONCLUSION: The visualization of NVC, nerve atrophy, and nerve dislocation on magnetic resonance imaging scans was not associated with pain outcome. A large vessel compressing the nerve and deforming the brainstem and proximal NVC were associated with a lesser pain control.