Changing role for preoperative embolisation in the management of arteriovenous malformations of the brain

BACKGROUND The aim of this study was to analyse the results of the use of preoperative embolisation in the management of arteriovenous malformations of the brain at one institution between 1989 and 1999. METHODS Two hundred and fifty consecutive cases of angiographically confirmed arteriovenous malformations underwent surgery by one surgeon. Cases of dural or spinal arteriovenous malformations have been excluded. Forty-five cases underwent preoperative embolisation. Embolisation was mostly by particulate embolic material delivered 4 to 6 days before the intended surgery. The incidence of embolisation declined from 21 cases of the first 50 arteriovenous malformation cases surgically treated to five in the last 50 cases. For arteriovenous malformations of less than 3 cm, only the first two temporal quintiles had embolised cases; six in the first and three in the second. Outcome was measured by the Modified Rankin Scale. RESULTS By 12 months (or last follow up, if less than this time has elapsed) following surgery, 1.6% of patients had died, 2.4% had a Modified Rankin scale score of 4 or 5, 6.4% had a Modified Rankin scale score of 3, 8.4% had a Modified Rankin scale score of 2, 14.4% had a Modified Rankin scale score of 1, and 66.8% were without neurological deficit. There was no difference in outcomes in each of the temporal quintiles. The four deaths were related to intraoperative haemorrhage, ruptured aneurysm, acute myocardial infarction or unrelated infection. Angiographic cure was achieved in 244 of 246 surviving cases. The two cases with residual arteriovenous malformations underwent focussed irradiation. Permanent morbidity could be attributable to embolisation, intraoperative events (resection functional brain, arteriovenous malformation rupture, aneurysm rupture or myocardial infarction) or postoperative events (arterio-capillary-venous hypertensive syndrome or infection). Of these 29 patients 14 had undergone embolisation. Mortality and major morbidity (Modified Rankin scale score greater than 2 due to treatment) occurred in 8.8% undergoing embolisation compared with 1.9% not embolised. The cause for major morbidity in these four embolised cases was intraoperative or postoperative haemorrhage. CONCLUSIONS These results reflect that cases selected for embolisation were those at most risk from intraoperative haemorrhage. Arteriovenous malformations that are less than 3 cm in maximal diameter should only rarely be considered for preoperative embolisation because of their low surgical morbidity. In the presence of a significant deep perforating artery contribution that cannot be effectively embolised the risks of operative haemorrhage is high irrespective of the effectiveness of embolising ancillary non-perforating arteries.