Back

Redefining ischemic core, penumbra, and target mismatch on CT perfusion in acute anterior distal medium vessel occlusion

Cai, L. Y.; Hoseinyazdi, M.; Lakhani, D. A.; Salim, H.; Mei, J.; Dmytriw, A. A.; Guenego, A.; Nguyen, T. N.; Majmundar, S. C.; Leigh, R.; Marsh, E. B.; Llinas, R. H.; Urrutia, V. C.; Hillis, A. E.; Fiehler, J.; Albers, G. W.; Heit, J. J.; Faizy, T. D.; Yedavalli, V. S.

2025-03-26 neurology
10.1101/2025.03.25.25324574 medRxiv
Show abstract

BackgroundRecent trials of endovascular thrombectomy (EVT) for acute distal medium vessel occlusions (DMVOs) were negative but also used inconsistent imaging-based inclusion criteria, whereas many successful large vessel occlusion (LVO) EVT trials used empirically validated CT perfusion-based target mismatch (TMM) criteria: an ischemic penumbra (time-to-maximum [Tmax] >6s) to core (relative cerebral blood flow [rCBF] <30%) mismatch ratio (MMR) [&ge;]1.8 and volume (MMV) [&ge;]15mL. We aimed to determine optimal corresponding definitions in DMVOs to improve patient selection for EVT. MethodsWe retrospectively analyzed patients with acute anterior DMVOs from prospectively collected databases at four comprehensive stroke centers. To assess core, we evaluated how well pretreatment rCBF <20%, <30%, <34%, and <38% volumes correlated with MRI-based post-treatment follow-up volumes (FIVs) in successfully recanalized patients. To evaluate penumbra, we assessed how well pretreatment Tmax >4s, >6s, >8s, and >10s volumes correlated with FIVs in unrecanalized patients. Then, we evaluated whether these improved parameters for core and penumbra better quantified LVO TMM and identified an optimal DMVO TMM definition. ResultsIn 122 core patients, rCBF <38% most strongly correlated with FIVs (concordance correlation coefficient [CCC] 0.30 [95% CI, 0.15-0.48]), outperforming rCBF <30% (CCC 0.21 [0.10-0.35]) (p < 0.001). In 70 penumbra patients, Tmax >8s most strongly correlated with FIVs (CCC 0.49 [0.25-0.77]), outperforming Tmax >6s (CCC 0.39 [0.17-0.68]) (p < 0.001). In 180 patients undergoing EVT with Tmax >6s to rCBF <30% MMR [&ge;]1.8 and MMV [&ge;]15mL, recomputing MMR and MMV using Tmax >8s and rCBF <38% further separated those with favorable outcomes (p = 0.007). In the same cohort, Tmax >8s to rCBF <38% MMR [&ge;]2.2 and MMV [&ge;]10mL maximized the number of patients benefitted (p < 0.001, absolute risk reduction 26%). ConclusionsIn acute anterior DMVOs, rCBF <38% and Tmax >8s best correspond to ischemic core and penumbra, respectively; more favorably quantify LVO TMM; and reveal optimal TMM criteria. These results should be prospectively investigated as inclusion criteria for EVT in this population and suggest recent negative DMVO EVT trials may have been confounded by suboptimal patient selection.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.