Logo image
Back
ABSTRACT NUMBER: ESOC2026A1999 EFFECT OF INTRAVENOUS THROMBOLYSIS ACCORDING TO AN AUTOMATED PREDICTED CORE VOLUME ON NCCT IN THE EXTENDED TIME WINDOW
Abstract   Peer reviewed

ABSTRACT NUMBER: ESOC2026A1999 EFFECT OF INTRAVENOUS THROMBOLYSIS ACCORDING TO AN AUTOMATED PREDICTED CORE VOLUME ON NCCT IN THE EXTENDED TIME WINDOW

Alan Flores, Adrián Valls-Carbó, Santiago Ortega-Gutierrez, Natalia Pérez de la Ossa, Mikel Terceño, Pere Cardona, Francisco Purroy, Xabier Urra, Marc Ribó and María Hernández-Pérez
European stroke journal, Vol.11(Suppl 1), pp.i82-i83
05/06/2026
DOI: 10.1093/esj/aakag023.132

View Online

Abstract

Background and aims We explored whether an automated predicted core volume (aCore) on non-contrast CT (NCCT) using a deep-learning algorithm modifies the benefit and safety of intravenous-thrombolysis (IVT) in the extended time-window. Methods This multicenter study pooled two cohorts of consecutive stroke patients arriving 4.5–24 hours from onset (Jan-2019 to Dec-2021). IVT patients were treated per CTP-EXTEND criteria. aCore was later evaluated on NCCT using a validated algorithm (Methinks, Spain). Inverse probability weighting was adjusted for age, baseline NIHSS, occlusion-status, sex, and onset-to-arrival time. Interaction models (IVT × aCore) assessed mRS-shift at 3-months, favorable outcome (mRS 0–2), and symptomatic intracranial hemorrhage (sICH). Linear and spline-based interaction terms were evaluated, and bootstrapping (400resamples) was used to estimate probabilities of benefit or harm. Results Among 976 patients (mean age 74, median NIHSS 10, median onset-to-admission 605 minutes, mean aCore 12.5 mL), 155 received thrombolysis (36 bridging;IVT + EVT) and 826 received no-reperfusion. Favorable outcome occurred in 42% and sICH in 5%. Linear models showed a significant IVT-by-aCore interaction: each additional aCore mL increased IVT benefit by ~ 2% on mRS-shift (OR0.98, 95% CI0.97–1.00, P < 0.01) and 4% on favorable outcome (OR1.04, 95% CI1.02–1.05; P < 0.01). No interaction was observed for sICH. Spline analyses revealed a progressive benefit up to ~ 50 mL, followed by attenuation; sICH risk rose non-linearly beyond this range(Figures 1-2). Conclusions NCCT-based aCore significantly modifies the effect of IVT in the extended-window. Thresholds near 50 mL may define functional benefit and sICH risk, supporting its use as a pragmatic selection tool.
 Abstract

Details

Metrics

1 Record Views
Logo image