Complications from arteriovenous malformation radiosurgery: multivariate analysis and risk modeling

Abstract

Purpose/objective: To assess the relationships of radiosurgery treatment parameters to the development of complications from radiosurgery for arteriovenous malformations (AVM).

Methods and materials: We evaluated follow-up imaging and clinical data in 307 AVM patients who received gamma knife radiosurgery at the University of Pittsburgh between 1987 and 1993. All patients had regular clinical or imaging follow up for a minimum of 2 years (range: 24-96 months, median = 44 months).

Results: Post-radiosurgical imaging (PRI) changes developed in 30.5% of patients with regular follow-up magnetic resonance imaging, and were symptomatic in 10.7% of all patients at 7 years. PRI changes resolved within 3 years developed significantly less often (p = 0.0274) in patients with symptoms (52.8%) compared to asymptomatic patients (94.8%). The 7-year actuarial rate for developing persistent symptomatic PRI changes was 5.05%. Multivariate logistic regression modeling found that the 12 Gy volume was the only independent variable that correlated significantly with PRI changes (p < 0.0001) while symptomatic PRI changes were correlated with both 12 Gy volume (p = 0.0013) and AVM location (p = 0.0066).

Conclusion: Complications from AVM radiosurgery can be predicted with a statistical model relating the risks of developing symptomatic post-radiosurgical imaging changes to 12 Gy treatment volume and location.