Treatment outcomes for cranial dural arteriovenous fistulas by anatomical location: a location-stratified meta-analysis of Microsurgery, embolisation and radiosurgery.

[OBJECTIVE] Cranial dural arteriovenous fistulas (dAVFs) exhibit variable angioarchitecture and rupture risk, closely linked to anatomical location. Comparative data to guide first-line treatment selection remain limited. This systematic review and meta-analysis evaluated angiographic and functional outcomes of microsurgical, endovascular, and radiosurgical monotherapy for cranial dAVFs, stratified by anatomical site: anterior cranial fossa (ACF), transverse-sigmoid sinus (TSS), and cavernous sinus (CS).

[METHODS] A systematic search of MEDLINE, Embase, Emcare, and Scopus were conducted through November 2024 following PRISMA guidelines. Eligible studies published since 2004 reported on ≥20 patients treated with a single modality and included site-specific outcomes. The primary outcome was angiographic obliteration; the secondary outcome was post-treatment functional dependence. Risk of bias was assessed using the MINORS tool. Random-effects meta-analyses were performed using logit-transformed proportions with Hartung-Knapp adjustment.

[RESULTS] Seventy studies comprising 3622 patients were included (ACF, n = 263; TSS, n = 798; CS, n = 2561). For ACF dAVFs, microsurgery achieved higher obliteration rates (90 %; 95 % CI, 82-94 %) compared with embolisation (73 %; 95 % CI, 59-83 %; p = 0.0056). In TSS dAVFs, embolisation achieved 59 % obliteration (95 % CI, 51-67 %), while limited data on microsurgery showed higher success (86 %; 95 % CI, 81-90 %). In CS dAVFs, embolisation and radiosurgery yielded comparable obliteration rates (85 % vs. 78 %, p = 0.31). Functional dependence rates following embolisation were low across locations (range, 5-10 %). Overall complication rates were 10 % for embolisation, 15 % for microsurgery, and 9 % for radiosurgery. Embolisation efficacy improved in studies published from 2016 to 2024 (p = 0.0440).

[CONCLUSION] Treatment outcomes for cranial dAVFs are highly dependent on anatomical location. Microsurgery provides superior efficacy for ACF lesions, while embolisation remains effective and preferred for TSS and CS dAVFs. Radiosurgery may be considered for selected CS lesions with unfavourable access. These findings support a location-based, multidisciplinary treatment algorithm.