Letter: Stereotactic Radiosurgery for Dural Arteriovenous Fistulas: A Systematic Review and Meta-Analysis and International Stereotactic Radiosurgery Society Practice Guidelines.

Abstract

To the Editor: We read with interest the article by Singh et al1 synthesizing the modern literature around the use of stereotactic radiosurgery (SRS) in the treatment of cranial dural arteriovenous fistulas (DAVFs). Neurointervention has disrupted the established treatment paradigm for many neurosurgical conditions. From intracranial aneurysms and large vessel occlusions to hydrocephalus, neuroendovascular surgery favorably altered the natural history and reduced the treatment-related morbidity of these and other disease processes. This has changed the mindset of clinicians who treat patients with cerebrovascular disease. For instance, a patient with an unruptured wide-necked aneurysm may be better served with an endovascular treatment that leaves a small neck remnant than with clip reconstruction that affords complete occlusion but requires a considerably more invasive surgical approach. Endovascular therapy has, in a generally safe and minimally invasive fashion, lowered the clinical risk of many pathologies without affording complete cure. In light of the available literature, a similar shift in mindset may be warranted regarding SRS for the treatment of cranial DAVFs. Just as SRS met with initial skepticism regarding its efficacy for vestibular schwannomas,2 neurointerventionalists and cerebrovascular surgeons have been slow to embrace its utility for DAVFs.3 In recent years, mounting evidence has emerged to support the efficacy of SRS as more than just a salvage therapy after failed endovascular embolization of DAVFs.4,5 In light of this, a recent systematic review and meta-analysis by Singh et al1 on behalf of the International Stereotactic Radiosurgery Society has summarized the DAVF SRS outcomes and provided some practice guidelines. In their analysis of 705 SRS-treated patients with DAVF in 21 studies, the authors reported complete occlusion and symptomatic cure rates of nearly 70% and 80%, respectively.1 In comparison, a recent meta-analysis of transarterial DAVF embolization using ethylene vinyl alcohol copolymer (Onyx; Medtronic Neurovascular) reported a complete occlusion rate of 82%.6 A multicenter retrospective analysis of the Consortium for Dural Arteriovenous Fistula Outcomes Research (CONDOR) database reported a complete obliteration rate of 55% after initial Onyx embolization.7 In this CONDOR study, the final obliteration rate with Onyx was less than 70%, even after all treatments. Surgical DAVF series typically report a higher obliteration rates than endovascular series, but many of these surgical cases have failed endovascular therapy, and operative morbidity and mortality rates up to 17% have been reported.3,8,9 The rates of post-SRS hemorrhage and permanent neurological deficits reported by Singh et al1 in their meta-analysis were 1.1% and 1.3%, respectively. By comparison, the risk of permanent neurological deficit after DAVF embolization was 4% in the meta-analysis by Sadeh-Gonike et al6 and 16.1% in the CONDOR database.7 In a separate analysis of unruptured high-grade DAVFs from CONDOR, post-treatment hemorrhage rates in the observation, endovascular, surgical, and SRS groups were 6.9%. 1.6%, 2.3%, and 0%, respectively.4 Although there were relatively few patients in the SRS group (n = 19), the risk of post-treatment hemorrhage seems to be similar among the various therapeutic modalities for unruptured DAVFs. DAVFs can be rewarding lesions to address for neurointerventionalists, as they provide a satisfying combination of technical skill, anatomic knowledge, and innovative approaches (ie, transvenous, direct puncture, combined surgical and endovascular). However, these cases can require hours of radiation exposure for the patient over potentially multiple embolization sessions, and there is a modest risk of DAVF recurrence even after initial angiographic cure. Based on analyses of the CONDOR database, obliteration of a low-grade DAVF after a single embolization session was achieved in <40% of cases,10 and the rate of recurrent arteriovenous shunting after angiographic cure (which required a mean of >2 embolization sessions) was 7%.11 Based on these findings, SRS may warrant consideration as a first-line treatment for unruptured DAVFs.1 If patients understand the potential risks associated with the latency period between treatment and efficacy, SRS should be discussed with unruptured patients with DAVF as a viable and reasonable alternative to embolization and surgical ligation. Definitive evidence from randomized comparative trials among the various treatment strategies for DAVFs is unlikely to be forthcoming for such a rare condition. Therefore, we will continue to rely on large multicenter databases, such as CONDOR, for our best available evidence regarding management outcomes and complications in contemporary neurovascular practice. The practice guidelines put forth by the ISRS underscore the favorable risk to benefit profile of SRS for appropriately selected DAVFs.