Purpose In the event of an acute ischemic stroke due to large vessel occlusion (AIS-LVO), patients with large core infarction and malignant edema have worse outcomes. Core infarction size growth is caused by poor cerebral blood flow and impaired microvascular perfusion. Cerebral microvascular perfusion is governed by the in-flow of arterial blood to the brain tissue, but also likely by the outflow of blood through the cerebral veins. Venous blood flow in the context of AIS-LVO may better indicate the the overall quality of tissue perfusion, as it reflects blood flow after passing the brain tissue. We determined if the venous microperfusion profile (VMP) predict ischemic lesion growth and final infarct core in AIS-LVO patients.

Materials and Methods We performed a multicenter, retrospective cohort study of AIS-LVO patients undergoing thrombectomy triage with CT angiography (CTA) and CT perfusion (CTP). Patients with motion artifact and incomplete electronical medical data were excluded. Patient details were obtained from prospectively maintained stroke databases and the electronic medical record. VMP was determined by opacification of the vein of Labbé, sphenoparietal sinus, and superficial middle cerebral vein on pre-thrombectomy CTA as: 0, not visible; 1, moderate opacification; and 2, full. Brain edema progression and infarct growth as assessed by Net Water Uptake (NWU), which was calculated on pre-treatment and post-thrombectomy non contrast computed tomography images using manual regions of interest. Primary outcome measure was ischemic lesion growth after thrombectomy. Secondary outcome was final core infarction volume, which was manually segmented on follow-up CT and MRI studies 24–48 hours after thrombectomy.

Results 250 patients met inclusion criteria. Median patient age was 76 (IQR 65–82). 50% were female. Linear regression models found that increased patient age (p=0.011), higher blood glucose levels (p=0.007), lower TICI scores (p<0.001) and reduced VMP (p<0.001) predicted increased core infarction growth (higher NWU). In a multivariate regression analysis, poor VMP predicted core infarct growth while controlling for age, blood glucose, and TICI score (β=-2.111840, [ 95% CI-2.808059256 -1.41562150]; p<001). In a secondary analysis focused on final infarct core, we excluded 5 patients due to missing final infarct size data. After controlling for age, blood glucose, and TICI score, poor VMP predicted higher final core infarct size (β= -22.57626,[ 95% CI -30.7799325 -14.3725813]; p<001).

Conclusion Poor cerebral perfusion on the venous microvascular profile predicts ischemic lesion growth and final infarct core volume in AIS-LVO patients treated with thrombectomy.

Disclosures T. Faizy: None. R. Kabiri: None. M. Leipzig: None. G. Broocks: None. S. Christensen: None. F. Flottmann: None. M. Lansberg: None. G. Albers: None. J. Fiehler: None. M. Wintermark: None. J. Heit: None.