Effect of treatment period on outcomes after stereotactic radiosurgery for brain arteriovenous malformations: an international multicenter study

Abstract

OBJECTIVE The role of and technique for stereotactic radiosurgery (SRS) in the management of arteriovenous malformations (AVMs) have evolved over the past four decades. The aim of this multicenter, retrospective cohort study was to compare the SRS outcomes of AVMs treated during different time periods. METHODS The authors selected patients with AVMs who underwent single-session SRS at 8 different centers from 1988 to 2014 with follow-up ≥ 6 months. The SRS eras were categorized as early (1988-2000) or modern (2001-2014). Statistical analyses were performed to compare the baseline characteristics and outcomes of the early versus modern SRS eras. Favorable outcome was defined as AVM obliteration, no post-SRS hemorrhage, and no permanently symptomatic radiation-induced changes (RICs). RESULTS The study cohort comprised 2248 patients with AVMs, including 1584 in the early and 664 in the modern SRS eras. AVMs in the early SRS era were significantly smaller (p < 0.001 for maximum diameter and volume), and they were treated with a significantly higher radiosurgical margin dose (p < 0.001). The obliteration rate was significantly higher in the early SRS era (65% vs 51%, p < 0.001), and earlier SRS treatment period was an independent predictor of obliteration in the multivariate analysis (p < 0.001). The rates of post-SRS hemorrhage and radiological, symptomatic, and permanent RICs were not significantly different between the two groups. Favorable outcome was achieved in a significantly higher proportion of patients in the early SRS era (61% vs 45%, p < 0.001), but the earlier SRS era was not statistically significant in the multivariate analysis (p = 0.470) with favorable outcome. CONCLUSIONS Despite considerable advances in SRS technology, refinement of AVM selection, and contemporary multimodality AVM treatment, the study failed to observe substantial improvements in SRS favorable outcomes or obliteration for patients with AVMs over time. Differences in baseline AVM characteristics and SRS treatment parameters may partially account for the significantly lower obliteration rates in the modern SRS era. However, improvements in patient selection and dose planning are necessary to optimize the utility of SRS in the contemporary management of AVMs.

Evaluation of the radiosurgical treatment of cerebral arteriovenous malformations: a retrospective single-center analysis of three decades

“…Consequently, in our cohort the modern era is characterized by smaller single treatment volumes but higher median prescription doses. Similar developments but a considerable increase in isocenters in the modern era have been described and are in line with our findings 17 18…”

“…In contrast, we found an overall hemorrhage rate after first GKRS of 9% of all patients, corresponding well with previously published series 17 18 23 24 26. Although the overall hemorrhage rate among the non-interventional ARUBA cohort was lower than that of our post-interventional rate, our follow-up was significantly longer.…”

“…Although the overall hemorrhage rate among the non-interventional ARUBA cohort was lower than that of our post-interventional rate, our follow-up was significantly longer. Consequently, the yearly hemorrhage rate of 1.3% in our observation period after first GKRS is in line with the recent literature and is significantly lower than that in the non-interventional arm of the ARUBA study (2.2%) 17 18 23 24 26. Of note, according to a meta-analysis of 3923 patients, untreated AVMs have an overall yearly hemorrhage rate of 3.0% 27…”

“…However, our significantly higher rate of post-radiosurgical hemorrhage in the pre-modern era differs from other results 17 18. The higher hemorrhage rate among patients treated in the pre-modern era in our study may be due to significantly higher SP classes among the pre-modern treatment group.…”

“…Cases were rated according to clinical and radiological assessment scores (table 1). 13–16 We divided our study sample into two eras, pre-modern and modern, as others have done previously 17 18…”