OBJECTIVE: We hypothesized that coexisting extranidal arterial aneurysms (EAs) would be associated with an increased risk of incident intracranial hemorrhage (ICH) from brain arteriovenous malformation (BAVM) rupture. METHODS: To determine the presence of EAs and compare the sources and locations of ICH, we retrospectively reviewed the computed tomographic, magnetic resonance imaging, and angiographic studies of patients who presented between 1990 and 1999. EAs were defined as saccular luminal dilations of the parent feeding vessels that were proximally flow-related (i.e., at the circle of Willis), distally flow-related (i.e., distal to the circle of Willis), and unrelated (i.e., in circulation distant from the BAVM). RESULTS: Of 314 BAVM patients, 138 (44%) presented with ICH. In the ICH group, 22 patients (16%) had aneurysmal ICH, 100 (72%) had BAVM ICH, and 16 (12%) had ICH from an indeterminate source. There were 61 patients with 1 or more EAs (29 patients with 42 flow-related proximal aneurysms, 39 patients with 48 flow-related distal aneurysms, and 10 patients with 20 unrelated aneurysms). Multivariate regression analysis revealed that ICH patients were more likely than non-ICH patients to have a coexisting EA (35 versus 13%; odds ratio = 3.9; 95% confidence interval, 2.1-7.5; P < 0.001), but this effect was not present when only BAVM-related ICH was considered (odds ratio = 0.3; 95% confidence interval, 0.1-1.0; P = 0.052). Other independent predictors of ICH included small AVM size (<3 cm), exclusively deep venous drainage, and the presence of an intranidal aneurysm. CONCLUSION: Clinical presentation with ICH was associated with EA aneurysms, but the association was due to aneurysmal rather than BAVM rupture, suggesting that EAs and the BAVM ICH risks may be considered as separate entities in future studies.
OBJECTIVE: We hypothesized that coexisting extranidal arterial aneurysms (EAs) would be associated with an increased risk of incident intracranial hemorrhage (ICH) from brain arteriovenous malformation (BAVM) rupture. METHODS: To determine the presence of EAs and compare the sources and locations of ICH, we retrospectively reviewed the computed tomographic, magnetic resonance imaging, and angiographic studies of patients who presented between 1990 and 1999. EAs were defined as saccular luminal dilations of the parent feeding vessels that were proximally flow-related (i.e., at the circle of Willis), distally flow-related (i.e., distal to the circle of Willis), and unrelated (i.e., in circulation distant from the BAVM). RESULTS: Of 314 BAVMpatients, 138 (44%) presented with ICH. In the ICH group, 22 patients (16%) had aneurysmal ICH, 100 (72%) had BAVMICH, and 16 (12%) had ICH from an indeterminate source. There were 61 patients with 1 or more EAs (29 patients with 42 flow-related proximal aneurysms, 39 patients with 48 flow-related distal aneurysms, and 10 patients with 20 unrelated aneurysms). Multivariate regression analysis revealed that ICHpatients were more likely than non-ICHpatients to have a coexisting EA (35 versus 13%; odds ratio = 3.9; 95% confidence interval, 2.1-7.5; P < 0.001), but this effect was not present when only BAVM-related ICH was considered (odds ratio = 0.3; 95% confidence interval, 0.1-1.0; P = 0.052). Other independent predictors of ICH included small AVM size (<3 cm), exclusively deep venous drainage, and the presence of an intranidal aneurysm. CONCLUSION: Clinical presentation with ICH was associated with EA aneurysms, but the association was due to aneurysmal rather than BAVMrupture, suggesting that EAs and the BAVMICH risks may be considered as separate entities in future studies.
Authors: Rene O Sanchez-Mejia; Michael W McDermott; Jeffery Tan; Helen Kim; William L Young; Michael T Lawton Journal: Neurosurgery Date: 2009-02 Impact factor: 4.654