Comparison of Perfusion Imaging Definitions of the No‐Reflow Phenomenon after Thrombectomy—What Is the Best Perfusion Imaging Definition?

Abstract

The no‐reflow phenomenon is a potential contributor to poor outcome despite successful thrombectomy. There are multiple proposed imaging‐based definitions of no‐reflow leading to wide variations in reported prevalence. We investigated the agreement between existing imaging definitions and compared the characteristics and outcomes of patients identified as having no‐reflow.MethodsWe performed an external validation of 4 existing published definitions of no‐reflow in thrombectomy patients with extended Thrombolysis in Cerebral Infarction scale 2c to 3 (eTICI2c‐3) angiographic reperfusion who underwent 24‐hour perfusion imaging from 2 international randomized controlled trials (EXTEND‐IA TNK part‐1 and 2) and a multicenter prospective observational study. Receiver‐operating‐characteristic and Bayesian‐information‐criterion (BIC) analyses were performed with the outcome variable being dependent‐or‐dead at 90‐days (modified Rankin Score [mRS] ≥3).ResultsOf 131 patients analyzed, the prevalence of no‐reflow significantly varied between definitions (0.8–22.1%; p < 0.001). There was poor agreement between definitions (kappa 5/6 comparisons <0.212). Among patients with no‐reflow according to at least 1 definition, there were significant differences between definitions in the intralesional interside differences in cerebral blood flow (CBF) (p = 0.006), cerebral blood volume (CBV) (p < 0.001), and mean‐transit‐time (MTT) (p = 0.005). No‐reflow defined by 3 definitions was associated with mRS ≥3 at 90 days. The definition of >15% CBV or CBF asymmetry was the only definition that improved model fit on BIC analysis (ΔBIC = −8.105) and demonstrated an association between no‐reflow and clinical outcome among patients with eTICI3 reperfusion.ConclusionsExisting imaging definitions of no‐reflow varied significantly in prevalence and post‐treatment perfusion imaging profile, potentially explaining the variable prevalence of no‐reflow reported in literature. The definition of >15% CBV or CBF asymmetry best discriminated for functional outcome at 90 days, including patients with eTICI3 reperfusion. ANN NEUROL 2024