Associations between clinicoradiological variables and onset-to-imaging time in acute ischemic stroke: A retrospective study

Background: Many acute ischemic stroke (AIS) patients have an unknown onset time. These patients require advanced imaging or are sometimes excluded from treatment. We investigated the associations between clinicoradiological variables, as assessed with non-contrast computed tomography (NCCT) and CT angiography (CTA), and onset-to-imaging time in AIS patients included in the MR CLEAN registry, an endovascular treatment (EVT) registry. Methods: We included 4003 AIS patients from the MR CLEAN Registry. Patients were classified as in the early (≤4.5 h; ≤6 h) or late (>4.5 h; >6 h) onset-to-imaging time windows. Univariable and multivariable logistic regression assessed associations between baseline clinicoradiological variables and onset-to-imaging time. Accuracy was evaluated using receiver operating characteristic (ROC) curve analyses. Results: In multivariable logistic regression, Alberta Stroke Program Early CT Score (ASPECTS) (OR 0.90, 95 % CI 0.84-0.97), presence of leukoaraiosis (OR 1.51, 95 % CI 1.19-1.92), and moderate (OR 2.53, 95 % CI 1.16-5.54) and good (OR 3.90, 95 % CI 1.76-8.66) collateral score were significantly associated with an >4.5 h onset-to-imaging time. ASPECTS (OR 0.89, 95 % CI 0.81-0.98), presence of leukoaraiosis (OR 1.55, 95 % CI 1.14-2.11), and a good (OR 3.31, 95 % CI 1.17-9.39) collateral score were significantly associated with an >6 h onset-to-imaging time. Area under the ROC curves was 0.63 for both the >4.5 h and >6 h time windows. Conclusion: Among AIS patients included in the MR CLEAN registry, lower ASPECTS, presence of leukoaraiosis and a higher collateral score are associated with a late onset-to-imaging time (>4.5 h; >6 h after stroke onset). The accuracy of the association between clinicoradiological variables and onset-to-imaging time is poor.