A historical analysis of single-stage gamma knife radiosurgical treatment for large arteriovenous malformations: evolution and outcomes

• Published in: Acta neurochirurgica Vol. 154; no. 3; pp. 383 - 394
• Main Authors: Nagy, Gábor, Rowe, Jeremy G., Radatz, Matthias W. R., Hodgson, Timothy J., Coley, Stuart C., Kemeny, Andras A.
• Format: Journal Article
• Language: English
• Published: Vienna Springer Vienna 01.03.2012; Springer Nature B.V
• Subjects: Adolescent; Adult; Age; Aged; Angiography; Child; Child, Preschool; Clinical Article; Embolization; Embolization, Therapeutic - methods; Evolution; Female; Hemorrhage; Humans; Interventional Radiology; Intracranial Arteriovenous Malformations - epidemiology; Intracranial Arteriovenous Malformations - surgery; Intracranial Arteriovenous Malformations - therapy; Magnetic resonance imaging; Male; Medicine; Medicine & Public Health; Middle Aged; Minimally Invasive Surgery; Morbidity; Neurology; Neuroradiology; Neurosurgery; Radiosurgery - adverse effects; Radiosurgery - methods; Reoperation - methods; Retrospective Studies; Side effects; Surgical Orthopedics; Treatment Outcome; Young Adult; Embolization; Haemorrhage; Arteriovenous malformation; Radiosurgery; Gamma knife
• ISSN: 0001-6268; 0942-0940
• DOI: 10.1007/s00701-011-1245-5

Background Large arteriovenous malformations (AVMs) remain challenging and difficult to treat, reflected by evolving strategies developed from simple radiosurgical plans, to encompass embolization and, recently, staged volume treatments. To establish a baseline for future practice, we reviewed our clinical experience. Method The outcomes for 492 patients (564 treatments) with AVMs >10 cm 3 treated by single-stage radiosurgery were retrospectively analysed in terms of planning, previous embolization and size. Results Twenty-eight percent of the patients presented with haemorrhage at a median age of 29 years (range: 2–75). From 1986 to 1993 (157 patients) plans were simplistic, based on angiography using a median of 2 isocentres and a marginal dose of 23 Gy covering 45-70% of the AVM (median volume 15.7 cm 3 ). From 1994 to 2000 (225 patients) plans became more sophisticated, a median of 5 isocentres was used, covering 64-95% of the AVM (14.6 cm 3 ), with a marginal dose of 21 Gy. Since 2000, MRI has been used with angiography to plan for 182 patients. Median isocentres increased to 7 with similar coverage (62-94%) of the AVM (14.3 cm 3 ) and marginal dose of 21 Gy. Twenty-seven percent, 30% and 52% of patients achieved obliteration at 4 years, respectively. The proportion of prior embolization increased from 9% to 44% during the study. Excluding the embolized patients, improvement in planning increased obliteration rates from 28% to 36% and finally 63%. Improving treatment plans did not significantly decrease the rate of persisting radiation-induced side effects (12–16.5%). Complication rate rose with increasing size. One hundred and twenty-three patients underwent a second radiosurgical treatment, with a 64% obliteration rate, and mild and rare complications (6%). Conclusions Better visualization of the nidus with multimodality imaging improved obliteration rates without changing morbidity. Our results support the view that prior embolization can make interpretation of the nidus more difficult, reducing obliteration rate. It will be important to see how results of staged volume radiosurgery compare with this historical material.