Hemorrhage Risk after Stereotactic Radiosurgery for Patients with Unruptured Brain Arteriovenous Malformation: Significance of Vascular Compactness

Abstract

<h3>Purpose/Objective(s)</h3> Our purpose was to investigate whether the morphology (i.e., compact/ diffuse) of brain arteriovenous malformations (bAVMs) correlates with the incidence of hemorrhagic events in patients with unruptured bAVMs after receiving non-invasive stereotactic radiosurgery (SRS). <h3>Materials/Methods</h3> This study retrospectively included 264 adult patients with unruptured bAVMs undergoing upfront SRS from multi-institutions. Two senior neurosurgeons confirmed the unruptured presentation. The follow-ups were performed in a 6-month interval. Hemorrhagic events were defined as bleeding signs on computed tomography (CT) or magnetic resonance images (MRI). The morphology of bAVMs was evaluated using fully automated segmentation, which calculated the proportion of vessel, brain tissue, and cerebrospinal fluid in bAVMs on T2-weighted MRI. Compactness index, defined as a ratio of vessel to brain tissue, was proposed and further dichotomized based on Youden indices to classify the bAVMs into compact or diffuse type. Cox proportional hazard model was used to identify the independent factors for the occurrence of post-SRS hemorrhage. <h3>Results</h3> The median follow-up duration was 62.1 months. Post-SRS hemorrhage occurred in 14 (5.3%) patients with 2 bAVMs bleeding twice, incurring an annual bleeding risk of 0.9%. Multivariable analysis with stepwise backward variable selection method revealed that bAVM morphology (compact versus diffuse), bAVM volume and prescribed margin dose were significant predictors for post-GKRS hemorrhage. The incidence density of post-GKRS hemorrhage in the compact and diffuse bAVMs were 4.9 and 11.8 (hemorrhagic events/ 1000 person-years (P-Ys)). Among the diffuse bAVMs (n=153), the post-GKRS hemorrhage rate was lower for smaller bAVM volume, <15ml versus 15-30ml versus >30ml (2.2 versus 13.5 versus 26.3 hemorrhagic events/ 1000 P-Ys; p=0.044) and for bAVMs receiving margin dose >16.6 Gy versus <16.6 Gy (5.2 versus 25.8 hemorrhagic events/ 1000 P-Ys; p=0.004). For the compact bAVMs (n=111), the post-GKRS hemorrhage rate was not significantly different between the dose and volume groups. <h3>Conclusion</h3> Compact and smaller bAVMs, with higher prescribed margin dose harbor lower risks of post-SRS hemorrhage. Among the diffuse bAVMs, post-GKRS hemorrhage risk varied with bAVM volume and margin dose. The finding could help guide GKRS treatment decision-making for the unruptured bAVMs.