Quantitative imaging outperforms No-reflow in predicting functional outcomes in a translational stroke model

Abstract

Microvascular dysfunction and no-reflow are considered a major cause of secondary damage despite revascularization in acute ischemic stroke (AIS), ultimately affecting patient outcomes. We used quantitative PET-MRI imaging to characterize early microvascular damages in a preclinical non-human primate model mimicking endovascular mechanical thrombectomy (EVT). During occlusion, PET perfusion and MRI diffusion were used to measure ischemic and lesion core volumes respectively. Following revascularization, multiparametric PET-MRI included perfusion, diffusion, blood-brain barrier (BBB) permeability MRI, and ¹⁵O-oxygen metabolism PET. Lesion growth on MRI was evaluated at one week, and the neurological score was assessed daily; a poor outcome was defined as a score>6 (0-normal, 60-death) after one week. Early after recanalization, the gold-standard PET ischemic threshold (<0.2 ​mL/min/g) identified post-EVT hypoperfusion in 67 ​% of the cases (14/21) located in the occlusion acute lesion. Acquired 110 ​min post-EVT, the area of MRI Tmax hypoperfusion was larger and even more frequent (18/20) and was also located within the acute lesion. Eight of the total cases (38 ​%) had a poor outcome, and all of them had no-reflow (7/8 MRI no-reflow and 6/8 ​PET no-reflow). Diffusion ADC alterations and post-EVT oxygen extraction fraction (OEF) values were significantly different in PET no-reflow cases compared to those without no-reflow, exhibiting an inverse correlation. Independently of no-reflow, long perfusion Tmax and post-EVT high BBB Ktrans in the lesion core were the hallmarks of poor outcome and infarct growth. This early quantitative imaging signature may predict infarct growth and poor outcome and help to identify neuroprotection targets.