ABSTRACT NUMBER: ESOC2026A1958 DOES HEMATOMA EVACUATION REDUCE RISK OF COGNITIVE DECLINE AFTER NON-TRAUMATIC SUPRATENTORIAL INTRACEREBRAL HEMORRHAGE: A MISTIE III ANALYSIS

Abstract Background and aims Nontraumatic supratentorial intracerebral hemorrhage (ICH) is associated with increased risk of progressive cognitive decline. Our aim was to test whether minimally invasive surgery (MIS) reduces risk of cognitive decline and whether ICH location influences this risk in a secondary analysis of the MISTIE-III trial. Methods The exposure was end of treatment (EOT) ICH volume. The primary outcomes were Mini Mental Status Exam (MMSE) scores at 1, 6, and 12 months. Secondary outcomes were change in MMSE score and cognitive impairment (MMSE<24) at one year. We evaluated the association between EOT volume and MMSE scores using linear and logistic regression and cubic spline modeling with multivariable risk adjustment. Results MMSE data was available in 419/499 patients. MMSE score increased significantly with time (P < 0.0001) by 5.8 points (95% CI: 4.9-6.6) between day 30 and 180 and by an additional 1.1 points (95% CI: 0.5-1.6) between day 180 and 365. At one year, 130/331 (39.3%) patients were cognitively impaired. Factors which negatively impacted MMSE in adjusted analyses were older age, male sex, lower Glasgow Coma Scale score, higher NIHSS score, deep ICH location, and higher EOT ICH volume. EOT ICH volumes of < 5.0 to < 10.0 mL increased the probability of MMSE ≥24 at one year (P = 0.02 and P = 0.04, respectively). Conclusions After large ICH, cognition improves but remains impaired in almost 40% at one year especially for deep ICH. Surgery lowers the risk of dementia, but only with significant hematoma removal to less than 5-10 mL residual ICH volume. Conflict of interest Dr. Ziai is supported by the NIH and serves as an Associate Editor of Neurocritical Care. Dr Shah has nothing to disclose. Dr. Awad is supported by the NIH and reports consulting fees from Neurelis, Inc. and Ovid Rx, unrelated to ICH. He has served as expert on medicolegal cases related to ICH. • Dr. Hanley is supported by the NIH (U01NS080824 and U24TR001609), and reports personal fees from Op2Lysis, personal fees from BrainScope and Neurotrope, and non-financial support from Genentech outside the submitted work. Dr. Flaherty is supported by the NIH.