Abstract Background and aims Brain frailty, reflected by structural brain changes on CT, has been linked to heterogeneity in stroke presentation and recovery. In ESCAPE-NA1, CT-based brain frailty markers were associated with higher baseline stroke severity and poorer 90-day outcomes after thrombectomy. We aimed to validate these associations in the HERMES collaboration and determine whether brain frailty modifies thrombectomy treatment effects. Methods We performed a pooled analysis of individual patient data from the HERMES collaboration. Brain frailty imaging markers were assessed on baseline non-contrast CT. Associations with baseline National Institutes of Health Stroke Scale (NIHSS) score and 90d-modified Rankin Scale (90d-mRS) were evaluated using multivariable regression, and interaction with thrombectomy effect analyzed. Results Among 1,730 patients in the pooled HERMES dataset, baseline NIHSS was higher in patients with white matter disease (WMD, adjusted β = 0.54, 95%CI:0.01-1.08) and medial temporal atrophy (adjusted β = 0.39, 95%CI 0.02-0.76). Greater brain frailty was associated with lower odds of favorable 90d-mRS, including for WMD (aOR 0.62, 95%CI 0.50–0.77), global cortical atrophy (GCA; aOR 0.65, 0.55–0.78), medial temporal atrophy (OR 0.70, 0.60–0.80), posterior atrophy (aOR 0.69, 0.58–0.83), and cerebral small vessel disease (CSVD) burden (aOR 0.74, 0.64–0.86). EVT treatment effect modification for 90d-mRS was observed for GCA (P = 0.036) and presence of lacunes (P = 0.034, Figure 1). Conclusions CT-based brain frailty markers are associated with higher baseline stroke severity and worse 90d functional outcome. Thrombectomy remained beneficial across brain frailty strata, but treatment benefit was smaller for patients with greater atrophy and CSVD burden, particularly lacunes. Conflict of interest All authors: nothing to disclose.
Bracard et al. (Fri,) studied this question.