Staged radiosurgical treatment for large benign cerebral lesions.

Object: Large brain lesions are considered unamenable to gamma knife treatment because of possible adverse effects. The proximity of eloquent brain to a target location, including some arteriovenous malformations (AVMs), still poses a challenge to the neurosurgeon. Because meningiomas and AVMs often show a slow or absent growth rate, partial radiosurgical treatment in cases with risk of unacceptable surgery-related morbidity is an alternative. The authors evaluated the results of staged treatment.

Methods: Since April 1992, 12 patients suffering from large benign tumors (meningiomas) and seven with AVMs considered inoperable underwent gamma knife radiosurgery in staged procedures. All patients with tumors underwent primary open surgery for partial removal of recurrent growth and subsequent radiosurgical treatment. Four men and eight women ranging in age from 26 to 73 years harbored tumors that ranged in volume from 19 to 90 cm3. Marginal dose ranged from 10 to 25 Gy prescribed in the 30 to 50% isodose. The time interval between the radiosurgical treatments varied from 1 to 8 months. Follow up ranged from 5 and 89 months. Three female and four male patients (age range 10-48 years) had undergone previous AVM embolization and were treated with margin doses of 18 to 22 Gy in the 40 to 50% isodose. The volume range of these AVMs was 11 to 25 cm3. The time interval between radiosurgical treatments ranged from 5 to 36 months. The mean follow up was 22.6 months. Neurological follow-up examinations showed clinical improvement in 11 patients (58%), whereas five (26%) remained unchanged. Among tumor patients no deterioration or further tumor growth was observed. Tumor necrosis demonstrated noticed in five patients (26%). One patient with an AVM suffered from seizures and hemiparesis. Follow-up magnetic resonance imaging of the residual nidus disclosed no rebleeding.

Conclusions: For selected patients with large benign lesions staged gamma knife radiosurgery could be an alternative, as evidenced by the authors' series in which there was no mortality and low incidence morbidity.