Imaging-to-recanalization delay influences perfusion CT threshold calibration for follow-up infarct volume estimation

Background and purpose Computed tomography perfusion (CTP) imaging is used in acute ischemic stroke (AIS) to estimate admission ischemic core volume (ICV). This retrospective single-center study evaluated how relative cerebral blood flow (rCBF) thresholds applied within a single postprocessing software package (syngo.via Neuro Perfusion) relate to follow-up infarct volume (FIV), with emphasis on the effect of imaging-to-recanalization delay. Patients and methods Among 555 consecutive patients undergoing endovascular treatment (EVT) for AIS between January 2016 and October 2021, 236 patients with middle cerebral artery or internal carotid artery occlusion were included. Admission ICV was calculated using rCBF thresholds from < 10% to < 30% in 2% increments and compared with FIV on postprocedural non-enhanced CT. Patients achieving complete reperfusion were further analyzed using cumulative 10-minute delay windows from ≤ 90 to ≤ 130 min. Results Complete reperfusion was achieved in 102 patients. Among patients achieving complete reperfusion within ≤ 100 min (n = 43), the highest observed correlation between admission CTP-derived ICV and FIV was seen at rCBF thresholds of < 22% and < 20% (R=0.86 for both). Between these tied thresholds, rCBF < 20% showed less frequent and smaller-volume ICV overestimation. In the broader subgroup with complete reperfusion within ≤ 120 min (n = 64), the highest observed correlation was seen at rCBF < 24% (R=0.84). Conclusions In this vendor-specific study, the best-correlating admission CTP-derived ischemic core threshold relative to follow-up infarct volume varied with imaging-to-recanalization delay. This pattern may be relevant when interpreting perfusion-derived core estimates in acute stroke workflows with variable non-avoidable treatment delays, but should be considered exploratory.