Object Arteriovenous malformations (AVM) of the corpus callosum were rare lesions, accounting for 9–11% of brain AVMs. Microsurgical resection of such lesions remained difficult regarding deep location and possible neuropsychological disasters of extended callosal resection. The introduction of endovascular and radiation therapies fundamentally changed the outcome of these lesions.

Methods Since 1995, clinical and angiographic data of cerebral AVMs were prospectively collected. We reviewed data from patients treated for an AVM of the corpus callosum and discussed factors influencing the endovascular approach of such lesions.

Results Thirty-eight patients (mean age: 31 year) were consecutively treated by endovascular techniques. 78.9% (30 cases) presented with intracranial haemorrhage. 15 AVMs (39.5%) were anterior, 18 (47.4%) posterior and 5 (13.1%) holocallosal. The Spetzler-Martin grade was I in 2 cases (5.2%), II in 20 (52.6%), III in 9 (23.7%), IV in 6 (15.8%) and V in 1 case (2.6%). The nidus was compact in 19 cases (50%), diffuse in 13 (34.2%) and multifocal in 6 (15.8%). Both anterior and posterior circulation branches fed 14 nidi (36.8%). The venous drainage was superficial in 3 cases (7.9%), deep in 28 (73.7%) and both in 7 (18.4%). 104 sessions have been performed with a procedural complication rate of 6.7%. The mean follow-up was 43.6 months with a last mRS <3 in 33 cases (86.8%). 22 patients (57.9%) have been totally cured. Univariate analysis of factors influencing the success of endovascular treatment showed that Spetzler-Martin grade ≥3 (p = 0.046), nidus >30 mm (p = 0.02), extension in eloquent area (p = 0.03) and holocallosal type (p0.005) significantly diminish chances to cure the AVM.

Conclusions AVMs of the corpus callosum seems to be difficult to treat with endovascular therapy alone. The goal of embolization should be the prevention of (re) bleeding and the decrease of nidus size. When endovascular treatment could not achieve to the complete cure of the AVM, associated radiotherapy permitted to increase the occlusion rate.

Disclosures T. Robert: None. R. Blanc: None. B. Gilboa: None. G. Ciccio: None. R. Fahed: None. H. Redjem: None. S. Pistocchi: None. B. Bartolini: None. M. Piotin: None.