Abstract
BACKGROUND: Venous outflow (VO) can reflect collateral blood flow, brain tissue perfusion, and clinical outcomes of acute ischemic stroke patients due to anterior circulation large vessel occlusion (AIS-LVO). However, there are currently only semi-quantitative scoring methods for VO. In this study, we aimed to explore a quantitative evaluation method for VO and investigate its relationship with tissue level collaterals (TLCs) and clinical outcomes. METHODS: AIS-LVO patients who experienced internal carotid artery (ICA) and/or the first segment of the middle cerebral artery (M1) occlusion and achieved successful recanalization after endovascular thrombectomy (EVT) were enrolled in this study. The attenuation values [Hounsfield unit (HU)] of the main outflow veins and dural venous sinuses were measured on computed tomography venography (CTV), namely the HU values of ischemic side vein of Trolard (ISVOT), ischemic side vein of Labbé (ISVOL), ischemic side superficial middle cerebral vein (ISSMCV), ischemic side internal cerebral vein (ISICV), and ischemic side transverse sinus (ISTS), as well as the HU values of normal side vein of Trolard (NSVOT), normal side vein of Labbé (NSVOL), normal side superficial middle cerebral vein (NSSMCV), normal side internal cerebral vein (NSICV), normal side transverse sinus (NSTS), and superior sagittal sinus (SSS), followed by calculating ratios as quantitative indicators of VO profiles. TLCs were measured on computed tomography perfusion (CTP) by the hypoperfusion intensity ratio (HIR). Clinical outcomes were measured by modified Rankin Scale (mRS) at 3 months after treatment. Multivariable regression analyses were used to determine the association of quantitative indicators with clinical outcomes. Pearson correlation was used to calculate the correlations between quantitative ratios and TLCs (HIR). RESULTS: A total of 104 AIS-LVO patients were finally enrolled in this study. The favorable outcome group (FOG) demonstrated significantly lower median age {63 [interquartile range (IQR), 51-76] vs. 75 (IQR, 67-81) years, P<0.001} and presentation National Institutes of Health Stroke Scale (NIHSS) [10 (IQR, 5-13) vs. 15 (IQR, 9-17), P=0.003] compared to the unfavorable outcome group (UOG). Besides, FOG showed higher median ISVOL/SSS [0.58 (IQR, 0.46-0.76) vs. 0.44 (IQR, 0.30-0.57), false discovery rate (FDR)-corrected P=0.011], ISVOL/ISTS [0.67 (IQR, 0.51-0.88) vs. 0.48 (IQR, 0.38-0.66), FDR-corrected P=0.009], ISSMCV/SSS [0.52 (IQR, 0.38-0.67) vs. 0.38 (IQR, 0.28-0.52), FDR-corrected P=0.022], ISSMCV/ISTS [0.57 (IQR, 0.46-0.77) vs. 0.45 (IQR, 0.33-0.64), FDR-corrected P=0.033], ISVOL/NSVOL [0.70 (IQR, 0.60-0.97) vs. 0.55 (IQR, 0.42-0.77), FDR-corrected P=0.022], ISSMCV/NSSMCV [0.73 (IQR, 0.56-0.88) vs. 0.58 (IQR, 0.40-0.75), FDR-corrected P=0.007] compared to UOG. Age [odds ratio (OR) =0.936; 95% confidence interval (CI): 0.901-0.972; P=0.007] and ISSMCV/NSSMCV (OR =15.608; 95% CI: 2.099-116.076; P=0.001) were independent predictors for FOG. The area under the curve (AUC) of predicted probability combining age and ISSMCV/NSSMCV reached 0.769 (cutoff value =0.701, sensitivity =0.442, specificity =0.981) when predicting FOG. The ratios of ISVOT/SSS (r=-0.515, P<0.001) and ISVOT/NSVOT (r=-0.519, P<0.001) showed moderate negative correlation with HIR. CONCLUSIONS: Quantitative VO profiles based on 4D-CTA could be a promising assessment of VO profiles in AIS-LVO patients.