Neurological morbidity and mortality associated with the endovascular treatment of cerebral arteriovenous malformations before and during the Onyx era

Abstract

OBJECT The widespread implementation of the embolic agent Onyx has changed the endovascular management of cerebral arteriovenous malformations (AVMs). Recent data suggest that outcomes following embolization and resection may have worsened in the Onyx era. It has been hypothesized that there may be increased complications with Onyx embolization and increased surgical aggressiveness in patients treated with Onyx. In this study the authors analyzed their institutional experience with the endovascular treatment of cerebral AVMs prior to and after the introduction of Onyx to determine factors associated with periprocedural neurological morbidity and mortality. METHODS A retrospective review was performed of all patients with cerebral AVMs undergoing embolization at the Barrow Neurological Institute from 1995 to 2012. RESULTS Endovascular treatment of 342 cerebral AVMs was performed over 446 treatment sessions (mean age 37.8 years, range 1–83 years). Clinical presentation included hemorrhage in 47.6%, seizures in 21.9%, headaches in 11.1%, and no symptoms in 10% of cases. The endovascular pretreatment strategy was preoperative in 78.9%, preradiosurgery in 9.1%, palliative in 5.3%, targeted in 4.4%, and curative in 2.3%. The median Spetzler-Martin grade was III. The mean number of arteries embolized was 3.5 (range 0–13 arteries), and the mean number of treatment sessions was 1.3 (range 1–4 sessions). Onyx was used in 105 AVMs (30.7%), and N-butyl cyanoacrylate (NBCA) without Onyx was used in 229 AVMs (67%). AVMs treated with Onyx had a higher mean number of arterial pedicles embolized than did NBCA cases (4.3 ± 2.7 vs 3.2 ± 2.4, respectively; p < 0.001) and a greater number of sessions (1.5 ± 0.7 vs 1.2 ± 0.5, respectively; p < 0.05). Unexpected immediate postprocedural permanent neurological deficits were present in 9.6% of AVMs, while transient deficits were present in 1.8%. There was 1 death (0.3%). Spetzler-Martin grade was not associated with differences in outcome, as permanent neurological deficits were observed in 12%, 9%, 13%, 11%, and 13% of AVMs for Spetzler-Martin Grades I–V, respectively (p = 0.91). The use of Onyx compared with NBCA was not associated with differences in periprocedural morbidity (p = 0.23). This lack of a difference persisted even when controlling for number of arteries and sessions (p = 0.14). Sex was not associated with differences in outcome. CONCLUSIONS Permanent and transient postprocedural neurological deficits were noted in 9.6% and 1.8% of all cases, respectively. AVM grade was not associated with endovascular outcome. Despite the greater number of sessions required and arteries embolized for Onyx cases, there was no statistically significant difference in the risk of neurological deficits following cerebral AVM embolization with Onyx and NBCA.