Abstract A147: Effect of Minimally Invasive Surgery for Spontaneous Intracerebral Hemorrhage on Aphasia and Dysarthria Outcomes: Post-hoc Analysis of MISTIE III

Background: Supratentorial intracerebral hemorrhage (ICH) frequently leads to disabling speech impairments such as aphasia and dysarthria. Most clinical trials of minimally invasive surgery (MIS) for ICH focus on gains in motor function with little information about recovery of language. This study investigated the trajectory and predictors of speech recovery in a phase 3 clinical trial of patients with severe ICH. Methods: We conducted a post-hoc analysis of all 499 participants from the Minimally Invasive Surgery Plus Alteplase for Intracerebral Haemorrhage Evacuation (MISTIE III) trial, of which 299 had complete NIHSS subscores for best language (Q9) and dysarthria (Q10) at days 30, 180, and 365. Recovery was defined as a subscore of 0 after impairment at baseline. The exposure was MIS. Primary outcomes were recovery of aphasia and dysarthria at one year. We used Kaplan–Meier analysis, and logistic regression models with stratification by lesion laterality, and end-of-treatment (EOT) ICH volume. Results: Median age was 61 years and 39.8% were female. Median [IQR] ICH volume was 43.6 [24] mL (62% deep location and 46% left hemisphere). By day 365, 61.4% of MISTIE and 63.9% of standard medical care patients recovered from aphasia, while dysarthria recovery occurred in 68.1% and 68.4%, respectively. Recovery was strongly influenced by lesion laterality: only 26.8% of patients with left hemisphere ICH recovered from aphasia compared to 85.2% with right hemisphere ICH (p < 0.001). Each 1 mL increase in EOT volume reduced the odds of aphasia recovery by 9% (OR 0.91, 95% CI 0.87–0.95, p < 0.001). Laterality was also the dominant predictor of dysarthria recovery (OR 0.17 for left-sided lesions, 95% CI 0.10–0.29, p < 0.001). MIS (55% of participants) was not independently associated with recovery in adjusted analyses, although trends toward earlier recovery were seen in survival analyses. Achieving an EOT volume <15 mL was associated with significantly higher rates of aphasia recovery (48.6% vs 38.2%, p = 0.038). Conclusions: Aphasia and dysarthria are common yet potentially reversible sequelae of ICH. Laterality and residual hematoma volume are the strongest predictors of recovery. MIS may facilitate earlier resolution but was not independently associated with one-year language recovery. These findings highlight the importance of volume reduction below 15 mL and consideration of language-specific endpoints in future ICH trials.