Symptomatic vasospasm and outcomes following aneurysmal subarachnoid hemorrhage: a comparison between surgical repair and endovascular coil occlusion

Abstract

OBJECT: The authors studied patients with aneurysmal subarachnoid hemorrhage (SAH) to determine whether the incidence of symptomatic vasospasm or overall clinical outcomes differed between patients treated with craniotomy and clip application and those treated by endovascular coil occlusion. METHODS: The authors reviewed 415 consecutive patients with aneurysmal SAH who had been treated with either craniotomy and clip application or endovascular coil occlusion at a single institution between 1990 and 2000. Three hundred thirty-nine patients underwent surgical clip application procedures, whereas 76 patients underwent endovascular coil occlusion. Symptomatic vasospasm occurred in 39% of patients treated with clip application, 30% of patients treated with endovascular coil occlusion, and 37% of patients overall. Compared with patients treated with clip application, patients treated with endovascular coil occlusion were more likely to suffer acute hydrocephalus (50 compared with 34%, p = 0.008) and were more likely to harbor aneurysms in the posterior circulation (53 compared with 20%, p < 0.001). Logistic regression models controlling for patient age, admission World Federation of Neurosurgical Societies (WFNS) grade, acute hydrocephalus, aneurysm location, and day of treatment revealed that, among patients with an admission WFNS grade of I to III, endovascular coil occlusion carried a lower risk of symptomatic vasospasm (odds ratio [OR] 0.34, 95% confidence interval [CI] 0.14-0.8) and death or permanent neurological deficit due to vasospasm (OR 0.28, 95% CI 0.08-1) compared with craniotomy and clip application. Similar models revealed no difference in the likelihood of a Glasgow Outcome Scale score of 3 or less at the longest follow-up review (median 6 months) between treatment groups (OR 0.58, 95% CI 0.28-1.21). CONCLUSIONS: Patients with better clinical grades (WFNS Grades I-III) at hospital admission were less likely to suffer symptomatic vasospasm when treated by endovascular coil occlusion, compared with craniotomy and clip application. Nevertheless, there was no significant difference in overall outcome at the longest follow-up examination between the two treatment groups.