Discrepancy between post-treatment infarct volume and 90-day outcome in the ESCAPE randomized controlled trial.

Ganesh, Aravind; Menon, Bijoy K; Assis, Zarina A; Demchuk, Andrew M; Al-Ajlan, Fahad S; Al-Mekhlafi, Mohammed A; Rempel, Jeremy L; Shuaib, Ashfaq; Baxter, Blaise W; Devlin, Thomas; Thornton, John; Williams, David; Poppe, Alexandre Y; Roy, Daniel; Krings, Timo; Casaubon, Leanne K; Kashani, Nima; Hill, Michael D; Goyal, Mayank

BACKGROUND Some patients with ischemic stroke have poor outcomes despite small infarcts after endovascular thrombectomy, while others with large infarcts sometimes fare better. AIMS We explored factors associated with such discrepancies between post-treatment infarct volume (PIV) and functional outcome. METHODS We identified patients with small PIV (volume ≤ 25th percentile) and large PIV (volume ≥ 75th percentile) on 24-48-h CT/MRI in the ESCAPE randomized-controlled trial. Demographics, comorbidities, baseline, and 24-48-h stroke severity (NIHSS), stroke location, treatment type, post-stroke complications, and other outcome scales like Barthel Index, and EQ-5D were compared between "discrepant cases" - those with 90-day modified Rankin Scale(mRS) ≤ 2 despite large PIV or mRS ≥ 3 despite small PIV - and "non-discrepant cases". Multi-variable logistic regression was used to identify pre-treatment and post-treatment factors associated with small-PIV/mRS ≥ 3 and large-PIV/mRS ≤ 2. Sensitivity analyses used different definitions of small/large PIV and good/poor outcome. RESULTS Among 315 patients, median PIV was 21 mL; 27/79 (34.2%) patients with PIV ≤ 7 mL (25th percentile) had mRS ≥ 3; 12/80 (15.0%) with PIV ≥ 72 mL (75th percentile) had mRS ≤ 2. Discrepant cases did not differ by CT versus MRI-based PIV ascertainment, or right versus left-hemisphere involvement (p = 0.39, p = 0.81, respectively, for PIV ≤ 7 mL/mRS ≥ 3). Pre-treatment factors independently associated with small-PIV/mRS ≥ 3 included older age (p = 0.010), cancer, and vascular risk-factors; post-treatment factors included 48-h NIHSS (p = 0.007) and post-stroke complications (p = 0.026). Absence of vascular risk-factors (p = 0.004), CT-based lentiform nucleus sparing (p = 0.002), lower 24-hour NIHSS (p = 0.001), and absence of complications (p = 0.013) were associated with large-PIV/mRS ≤ 2. Sensitivity analyses yielded similar results. CONCLUSIONS Discrepancies between functional ability and PIV are likely explained by differences in age, comorbidities, and post-stroke complications, emphasizing the need for high-quality post-thrombectomy stroke care. CLINICAL TRIAL REGISTRATION https://clinicaltrials.gov/ct2/show/NCT01778335 .