Document resource
Introduction Successful reperfusion (eTICI 2b-3) during thrombectomy treatment of ischemic stroke leads to improved outcomes, but approximately half of reperfused patients still suffer poor outcomes. One possible explanation for these poor outcomes is that eTICI does not capture heterogeneity in reperfusion across arterial territories. Differences in the distribution and eloquence of reperfused regions may drive this variability. A more nuanced topography-informed, regional assessment of reperfusion may serve as an adjunct to eTICI to improve outcome prediction after thrombectomy. We determined how eTICI and regional reperfusion impacted early predictors, rapid neurologic deterioration (RND) and rapid neurologic improvement (RNI), of functional outcomes.Methods We performed a post-hoc analysis of the prospective, NIH-funded CRISP 2 thrombectomy study that determined predictors of core growth during transfer. Patients who underwent thrombectomy for internal carotid or middle cerebral artery occlusion were included. Patient characteristics and thrombectomy details were captured in the study database. Regional TICI reperfusion patterns were categorized by topography: Class 1 (complete reperfusion in all territory), Class 2 (poor reperfusion in parietal/occipital lobe), Class 3 (poor reperfusion in frontal lobe), Class 4 (reperfusion limited to temporal lobe), and Class 5 (poor reperfusion in entire territory). Classes 4-5 were combined into an incomplete category due to limited sample size and served as the reference for odds ratios. Univariate and multivariate analyses compared the prognostic performance of regional TICI versus eTICI for RND and RNI, defined as 24-hour NIHSS increase ≥3 and decrease ≥3, respectively.Results Of 378 patients, 187 met inclusion criteria. RND occurred in 29 patients, and RNI occurred in 116 patients. In multivariable analysis adjusted for age and thrombolysis, regional TICI was associated with neurological deterioration, with odds ratios of 0.11 (CI 0.02-0.49; p < 0.01) for Class 1 and 0.25 (CI 0.07-0.86; p = 0.02) for Class 2, whereas Class 3 was not significantly associated (OR 0.35, CI 0.09-1.37; p = 0.13). Furthermore, regional TICI classification demonstrated modest improvement over eTICI in predicting rapid neurological deterioration (AUC 0.70 vs 0.68) while correctly reclassifying 24% of patients toward worse outcomes based on net reclassification index. In contrast, eTICI demonstrated better performance in predicting rapid neurological improvement (AUC 0.72 vs 0.71).Conclusion While eTICI predicts rapid neurological improvement after thrombectomy, regional TICI provides greater sensitivity for identifying patients at risk of neurological deterioration and poor outcomes. As an adjunct to eTICI, a topography-informed regional classification may improve outcome stratification among patients with successful reperfusion.Disclosures N. Krothapalli: 1; C; Research support from American Heart Association GWTG-Stroke Program. P. Konduri: 1; C; Funding from 2025-26 NIH StrokeNet Training Program (U10NS086487). 4; C; Cofounder and shareholder of inSteps BV. X. Li: None. J. Sim: None. G. Albers: 2; C; Compensation from Biogen, iSchemaView, and Genentech for consultant services. 4; C; Stock holdings in iSchemaView. M. Lansberg: 1; C; Grants from the National Institute of Neurological Disorders and Stroke. J. Heit: 2; C; Consulting fees from Medtronic and MicroVention. 4; C; Stock holdings in Dragon Vascular. 6; C; Member of the Medical and Scientific Advisory Board of iSchemaView.Abstract E-227 Figure 1-2