LAUSR

We would like to thank Su X, et al for their interest and their remarks on our research. We analyzed the natural history, angiographic characteristics and treatment outcomes of anterior cranial fossa dural arteriovenous fistulas (ACFdAVFs) during a period of 27 years. This analysis determined that microsurgery achieved ACFdAVF closure in all cases (35/35), while endovascular therapy (EVT) achieved closure in 53% (9/17) of cases. As we acknowledged in the limitations of our study, there has been a robust improvement in endovascular techniques in the last two decades and the inclusion of data acquired within 27 years may not reflect the most up to date EVT. The introduction of new liquid embolics, more navigable microcatheters and improved imaging quality allows the endovascular treatment of complex cerebrovascular lesions. 4 Two recent studies reported high occlusion rates of 83% (19/23) and 91% (31/34) in cohorts of ACFdAVFs treated endovascularly. 6 These studies have the inherent limitations of selfreport bias, no standardized imaging followup and lack of core laboratory adjudication. However, the high angiographic cure reported in these studies suggest that more effective EVT can achieve better outcomes. Nonetheless, other series have reported higher rates of angiographic cure with microsurgery: 95% (19/20), 100% (14/14), and 100% (11/11). Additionally, a metaanalysis of 81 patients with ACFdAVFs reported complete angiographic closure in all the patients (65) treated with microsurgery, and only 47% (15/32) in patients treated with EVT. There were no significant differences between EVT and microsurgery in the 30day incidence of strokes, transient ischemic attacks, visual deficits, newonset seizures and intracranial hemorrhages. The heterogeneity of study protocols, the small size of the cohorts included in the metaanalysis and the report bias of each center are inherent limitations of this analysis. A randomized study comparing both interventions would be logistically challenging due to the low incidence of ACFdAVFs, the heterogeneous and unique angioarchitecture of each lesion and the multiple possible treatment approaches of EVT. As the authors mentioned, a frontal craniotomy is more invasive and may lead to complications such as infections and cerebrospinal fluid leaks, among others. Additionally, retraction injury and damage to the olfactory track can occur. However, the endovascular route is not completely innocuous. A transarterial approach may jeopardize the ophthalmic artery and risk retinal infarction. A transvenous approach can be complicated with perforation of cortical veins and intracranial hemorrhage. This is especially troublesome as cortical venous drainage is a core feature of ACFdAVF. CONDOR’s data show that microsurgery achieves a more effective closure of the fistula with a better safety profile than EVT. A pretreatment diagnostic angiogram provides valuable information to characterize the angioarchitecture of the fistula and determine if a favorable endovascular route can be attempted. Short and straight cortical veins with a patent superior sagittal sinus, or a large ophthalmic artery, could ease a safe and effective EVT. Patients and family members should be informed of all the available options and the most comprehensive scientific evidence. In view of the lack of type I level of evidence, the best level of evidence on the natural history and treatment outcomes of ACFdAVFs comes from large multicenter case series such a CONDOR. As with other cerebrovascular pathologies that are not common, treatment decisions should also be guided by local contingencies, such as level of expertise and access to specialized care.