Stereotactic radiosurgery for symptomatic solitary cerebral cavernous malformations considered high risk for resection

J Neurosurg 113:23–29, 2010. DOI: 10.3171/2010.1.JNS081626

A retrospective study was conducted to reassess the benefit and safety of stereotactic radiosurgery (SRS) in patients with solitary cerebral cavernous malformations (CCMs) that bleed repeatedly and are poor candidates for surgical removal.

Methods. Between 1988 and 2005 at the University of Pittsburgh, the authors performed SRS in 103 evaluable patients (57 males and 46 females) with solitary symptomatic CCMs. The mean patient age was 39.3 years. Ninetyeight percent of these patients had experienced 2 or more hemorrhages associated with new neurological deficits. Seventeen patients (16.5%) had undergone attempted resection before radiosurgery. Ninety-three CCMs were located in deep brain structures and 10 were in subcortical lobar areas of functional brain importance. The median malformation volume was 1.31 ml, and the median tumor margin dose was 16 Gy.

Results. The follow-up ranged from 2 to 20 years. The annual hemorrhage rate—that is, a new neurological deficit associated with imaging evidence of a new hemorrhage—before SRS was 32.5%. After SRS 22 hemorrhages were observed within 2 years (10.8% annual hemorrhage rate) and 4 hemorrhages were observed after 2 years (1.06% annual hemorrhage rate). The risk of hemorrhage from a CCM was significantly reduced after radiosurgery (p < 0.0001). Overall, new neurological deficits due to adverse radiation effects following SRS developed in 14 patients (13.5%), with most occurring early in our experience. Modifications in technique (treatment volume within the T2- weighted MR imaging–defined margin, use of MR imaging, and dose reduction for CCM in critical brainstem locations) further reduced risks after SRS.

Conclusions. Data in this study provide further evidence that SRS is a relatively safe procedure that reduces the rebleeding rate for CCMs located in high-surgical-risk areas of the brain.