Impact of brain frailty on language recovery in patients with acute post-stroke aphasia: a post-hoc analysis of the LEXI randomized controlled trial

Abstract Background The influence of brain frailty on post-stroke aphasia recovery in the acute phase and its interaction with Speech and Language Therapy (SLT) intensity is unclear. We investigated the association between brain frailty components and language outcome and assessed whether brain frailty modifies the dose-dependent treatment effect of SLT. Methods This is a post-hoc analysis of the LEXI multicenter randomized controlled trial, that enrolled patients with acute post-stroke aphasia from 07/2021 to 09/2024. Patients with pre-stroke cognitive impairment or dementia were not eligible for the trial. Participants were randomized to tablet-assisted SLT (Neolexon application) or standard SLT. Brain frailty was retrospectively assessed by outcome-blinded raters on interrater-adjusted baseline non-contrast CT, using cortical and subcortical atrophy, white matter changes (Fazekas score), lacunes, and chronic infarctions to derive a composite brain frailty score (BFS; range 0–3; higher scores indicate higher brain frailty burden). Primary outcome was the 90-days Bielefelder Aphasia Screening Test percentile rank. Results 56 patients (median age 75 years; 48.2% female) were included. 43 patients (76.8%) had a BFS of 0–1 and 13 (23.2%) of 2–3. Higher BFS were independently associated with worse 90-day language outcome (β -5.3; 95%CI -10.35 to -0.21; p = 0.042), with higher Fazekas scores contributing most (β -4.0; 95%CI -7.6 to -0.5; p = 0.028). Higher SLT dose was associated with improved outcome (β 0.34; 95%CI 0.07 to 0.61; p = 0.016). There was no statistical evidence that brain frailty modified the association between SLT duration and language outcome (p interaction = 0.507), although the interaction analysis was limited by sample size. Conclusions Higher white matter disease burden, as a component of brain frailty, was associated with poorer post-stroke language recovery in patients without pre-stroke cognitive impairment or dementia. We found no evidence that brain frailty modified the association between SLT intensity and language outcome, although larger studies are needed to assess effect modification. Routine imaging-based assessment of white matter disease burden may improve prognostic stratification and neurorehabilitation trial design. Clinical Trial Registration ClinicalTrials.gov Identifier: NCT04080817; Study Details. Neolexon® Aphasia-App in Acute Aphasia After Stroke. ClinicalTrials.gov. Date of Registration: September 4, 2019)