Dear Sir:

Endovascular treatment (EVT) has become a routine therapeutic approach for acute ischemic stroke (AIS) patients with large vessel occlusion, emphasizing the importance of successful arterial reperfusion as a predictor of positive clinical outcomes. However, a significant subset of patients experiences poor functional outcomes despite achieving complete angiographic reperfusion. Factors contributing to this disparity include age, brain atrophy, pre-EVT National Institutes of Health Stroke Scale (NIHSS), large infarct core at admission, poor arterial collaterals, and the number of passes during the EVT procedure [1].

While the majority of AIS studies focus on the arterial system, some recent studies have shown an association between cortical venous drainage and clinical outcomes in AIS patients [2-4]. These studies included patients who underwent different treatments and used diverse methodologies. However, it is still unclear the impact of venous drainage delay in patients who undergo EVT and whether this delay persists after the procedure. We aimed to evaluate the predictive value of cortical venous delay, measured through dynamic magnetic resonance angiography (dMRA)—a time-resolved sequence capturing arterial filling and venous emptying—before and after EVT, on infarct volume and functional outcomes in patients with large anterior artery occlusion undergoing EVT.

This project is a preplanned substudy of the FUtile Recanalization in Ischemic Acute Stroke (FURIAS) study. The study design and methodology have been published [5]. The magnetic resonance imaging (MRI) protocol has been described in Supplementary Table 1. Briefly, we prospectively enrolled patients with acute intracranial carotid artery or middle cerebral artery M1 or proximal M2 occlusion undergoing EVT that received serial brain MRI. The Research Ethics Committee of the Germans Trias Hospital approved this study (approval number PI-15-071). Written informed consent was obtained from all patients. The brain MRI scans were systematically analyzed at three distinct time points: at hospital admission (pre-EVT), within 2 hours after EVT (postEVT), and 5 days after EVT by a trained investigator. A previous paper showed that the delay in cortical veins seems to be a stronger predictor of outcome than the delay in the deep veins [3]. Therefore, we decided to restrict our study to the cortical veins. Cortical venous delay was quantified by identifying the time point in which venous filling was maximal in the main cortical veins for both hemispheres and calculating the percentage of delay for each vein for both pre-EVT and post-EVT MRIs (Supplementary Figure 1). We considered successful reperfusion as a final modified Thrombolysis in Cerebral Infarction (mTICI) grade ≥2B. Associations between venous delay and infarct volumes post-EVT and on day 5 were studied using multivariable linear regression. An ordinal logistic regression was performed to study the influence of venous delay on the modified Rankin Scale (mRS) at 3 months. We did these analyses for the whole cohort and for the group of patients that achieved successful reperfusion. The analyses were adjusted by relevant variables according to previous literature (Supplementary Table 2).

We enrolled 94 stroke patients (mean age: 69.9; median NIHSS: 17; final mTICI ≥2b: 86.2%) undergoing EVT (the selection flowchart and further comparison are published elsewhere). From those, we could evaluate 89 pre-EVT dMRAs and 88 post-EVT dMRAs (Supplementary Figure 2). Eighty-two of them had evaluable pre-EVT and post-EVT dMRA. An overview of the patient sample characteristics is shown in Table 1 and Supplementary Table 3. Pre-EVT and post-EVT mean venous delays were 31.9% and 13.5%, respectively. Successful reperfusion resulted in a 15.5% reduction in post-EVT venous delay with no significant change for non-reperfused patients (Supplementary Figure 3). Pre-EVT venous delay showed no significant correlation with infarct volume or clinical outcome, adjusting for reperfusion success (Table 2). However, for each 10% increment in post-EVT venous delay, there was an adjusted 0.12 mL increase (95% confidence interval [CI], 0.05–0.19) in infarct volume at 5 days and higher odds of an elevated mRS score at 3 months (adjusted crude odds ratio 1.14; 95% CI, 1.01–1.29) (Table 2).

Our key finding was that post-EVT cortical venous delay was linked to larger infarct volume at 5 days and worse functional outcomes at 3 months. However, the association between preEVT venous delay and clinical and radiological outcomes was not significant after adjustments by collateral grade and reperfusion status. Previous studies have yielded mixed results in this respect [2-4,6-8]. Of note, venous delay in AIS can be explained by arterial occlusion, poor collateral flow, breakdown of the blood-brain barrier causing edema and venous compression, and microthrombi formation [9].

It is unclear whether our lack of significance is due to our small sample size or if reperfusion status and collateral grade might act as confounding variables, influencing radiological and clinical outcomes more than pre-EVT venous delay itself.

Notably, the study identified a trend for improved venous outflow post-EVT in successfully reperfused patients. Nevertheless, this group of patients still had some grade of venous delay, possibly due to tissue edema or microthrombi. While venous delay post-EVT was linked to larger infarct volumes, the association with clinical outcomes in the reperfused group was not significant, potentially due to limited statistical power.

Strengths of the study included systematic evaluation of veins using consistent radiological methods, quantitative assessment of venous delay, and dMRA allowing precise identification of maximal contrast enhancement. However, limitations included a small sample size, incomplete brain coverage in the imaging protocol, and the simplification of venous delay measures.

This study emphasized the need for further research on venous patterns in AIS, advocating for larger patient populations and using dynamic imaging techniques to better understand cerebrovascular circulation complexity. Our findings suggest that venous circulation plays a role in post-stroke recovery, highlighting the importance of comprehending venous function in ischemic stroke for identifying potential neuroprotective targets and improving outcomes.