Original Article
Post-thrombolysis computed tomography angiography and perfusion features associated with an unfavorable early neurological response: a single-center retrospective study in a clinically selected anterior-circulation cohort
Abstract
Background: Early neurological non-response or worsening after intravenous thrombolysis (IVT) for acute ischemic stroke (AIS) is associated with poor outcomes, but its post-treatment imaging correlates remain incompletely understood. This study evaluated computed tomography angiography (CTA) and computed tomography perfusion (CTP) features associated with an unfavorable early neurological response (UENR).
Methods: In this single-center retrospective study, consecutive patients with AIS who received IVT between 2020 and 2025 and underwent CTA and perfusion within 24 hours because of a lack of improvement or initially severe deficits were included. A UENR was defined as an improvement of no more than 1 point or any increase in the National Institutes of Health Stroke Scale (NIHSS) score at 24 hours. Patients with an improvement of at least 4 points or a 24-hour score of 0–1 comprised the favorable response (FR) group. Imaging variables included post-treatment vessel patency, collateral status assessed with the Tan score, and estimated ischemic core and total hypoperfusion volumes.
Results: Among 185 patients, 62 (33.5%) had an unfavorable response and 123 (66.5%) had a FR. The unfavorable-response group had a higher baseline NIHSS score (median 16 vs. 12; P<0.001) and a higher rate of poor 90-day functional outcome. Post-treatment imaging showed a lower vessel patency rate (24.2% vs. 78.9%; P<0.001), a higher prevalence of poor collateral status (66.1% vs. 17.1%; P<0.001), and a larger estimated ischemic core (median 28.5 vs. 6.1 mL; P<0.001) in the unfavorable-response group. In multivariable logistic regression, persistent occlusion [adjusted odds ratio (aOR), 7.85; 95% confidence interval (CI): 3.60–16.80], poor collateral status (aOR, 4.05; 95% CI: 1.85–8.60), and larger ischemic core volume per 10 mL increase (aOR, 1.82; 95% CI: 1.35–2.45) were independently associated with an unfavorable response (all P<0.001).
Conclusions: In this clinically selected cohort, a UENR was associated with persistent vessel occlusion, poor collateral status, and a larger estimated ischemic core on post-treatment imaging. These findings characterize the concurrent vascular and tissue state of patients with early non-response or worsening and should not be interpreted as baseline predictors.

