Improved cognitive outcomes with endovascular coiling of ruptured intracranial aneurysms: neuropsychological outcomes from the International Subarachnoid Aneurysm Trial (ISAT).

BACKGROUND AND PURPOSE The International Subarachnoid Aneurysm Trial (ISAT) reported lower rates of death and disability with endovascular versus neurosurgical treatment of ruptured intracranial aneurysms. However, assessment of functional outcome was limited to the modified Rankin Scale, which is known to be insensitive to cognitive function. A neuropsychological substudy (N-ISAT) was therefore done in all recruits from 8 ISAT centers in the United Kingdom. METHODS Detailed neuropsychological assessment was performed at a 12-month follow-up visit. Impairment was defined as performance below the 5th percentile of the study population on at least 2 tests in >or=2 major cognitive domains. Analysis was restricted to patients who were not known to be otherwise disabled according to the modified Rankin Scale (ie, modified Rankin Scale 0 to 2). RESULTS Of 836 patients randomized in ISAT in the 8 UK centers (411 allocated endovascular treatment versus 425 neurosurgery), 224 were dead or disabled before 12-month follow-up (78 allocated endovascular treatment versus 135 neurosurgery). Of the remaining 612 patients eligible for neuropsychological assessment, 137 (65 allocated endovascular treatment versus 72 neurosurgery) did not attend. Of the 474 nondisabled patients who were assessed, 152 (32.1%) had cognitive impairment. Patients with cognitive impairment had reduced self-reported health-related quality of life (P<0.001) in both treatment groups, but cognitive impairment was less common in those allocated endovascular treatment (70 of 262 versus 82 of 212 allocated neurosurgery, OR=0.58, 95% CI 0.38 to 0.87, P=0.0055). The incidence of epilepsy was also lower in the N-ISAT endovascular group (7 versus 18, OR=0.30, 0.11 to 0.77, P=0.005) but was independent of the effect on cognitive function. CONCLUSIONS Cognitive impairment occurred in approximately one third of patients who were not otherwise disabled according to the modified Rankin Scale in N-ISAT and was more frequent in the neurosurgery group. These results have implications for management of ruptured intracranial aneurysms and more generally for interpretation of the outcomes of clinical trials that use the modified Rankin Scale.