OBJECTIVE: In this paper the authors' goal was to define the long-term benefits and risks of stereotactic radiosurgery (SRS) for patients with arteriovenous malformations (AVMs) who underwent prior embolization.
METHODS: Between 1987 and 2006, the authors performed Gamma Knife surgery in 996 patients with brain AVMs; 120 patients underwent embolization followed by SRS. In this series, 64 patients (53%) had at least one prior hemorrhage. The median number of embolizations varied from 1 to 5. The median target volume was 6.6 cm3 (range 0.2–26.3 cm3). The median margin dose was 18 Gy (range 13.5–25 Gy).
RESULTS: After embolization, 25 patients (21%) developed symptomatic neurological deficits. The overall rates of total obliteration documented by either angiography or MRI were 35%, 53%, 55%, and 59% at 3, 4, 5, and 10 years, respectively. Factors associated with a higher rate of AVM obliteration were smaller target volume, smaller maximum diameter, higher margin dose, timing of embolization during the most recent 10-year period (1997–2006), and lower Pollock-Flickinger score. Nine patients (8%) had a hemorrhage during the latency period, and 7 patients died of hemorrhage. The actuarial rates of AVM hemorrhage after SRS were 0.8%, 3.5%, 5.4%, 7.7%, and 7.7% at 1, 2, 3, 5, and 10 years, respectively. The overall annual hemorrhage rate was 2.7%. Factors associated with a higher risk of hemorrhage after SRS were a larger target volume and a larger number of prior hemorrhages. Permanent neurological deficits due to adverse radiation effects (AREs) developed in 3 patients (2.5%) after SRS, and 1 patient had delayed cyst formation 210 months after SRS. No patient died of AREs. A larger 12-Gy volume was associated with higher risk of symptomatic AREs. Using a case-control matched approach, the authors found that patients who underwent embolization prior to SRS had a lower rate of total obliteration (p = 0.028) than patients who had not undergone embolization.
CONCLUSIONS: In this 20-year experience, the authors found that prior embolization reduced the rate of total obliteration after SRS, and that the risks of hemorrhage during the latency period were not affected by prior embolization. For patients who underwent embolization to volumes smaller than 8 cm3, success was significantly improved. A margin dose of 18 Gy or more also improved success. In the future, the role of embolization after SRS should be explored.