Background

Venous outflow (VO) has emerged as an important imaging biomarker for prognosticating outcomes in patients with acute ischemic stroke due to large vessel occlusion (AIS-LVO). Research has shown that unfavorable VO profiles are associated with a higher risk of cerebral edema,^1-3 progression of edema,^{4, 5} hemorrhagic transformation,⁶ and poor clinical outcomes,^{1, 7-10} even after successful reperfusion efforts. Conversely, favorable VO has been shown to predict better clinical outcomes, particularly in patients undergoing intravenous thrombolysis or mechanical thrombectomy.^11-13

Prolonged venous transit (PVT) is defined as the presence of a time to maximum (Tmax) of ≥10 seconds in the superior sagittal sinus (SSS), the torcula, or both. PVT serves as a surrogate marker for VO status, integrating information from both superficial and deep cerebral venous drainage systems. The maker is readily interpretable on post-processed Tmax maps from CT perfusion imaging. PVT has demonstrated utility by showing an association with unfavorable 90-day mRS in successfully reperfused AIS-LVO patients.^{14, 15} However, its prognostic significance concerning changes in neurological deficits over time from baseline has not been thoroughly investigated.

The National Institutes of Health Stroke Scale (NIHSS) is commonly used to quantify stroke deficits over time, starting from presentation. It offers insight beyond standard functional measures by tracking neurological improvement as an outcome. It is considered a reliable predictor of hospital length of stay, functional outcome at discharge, discharge disposition, and in-hospital mortality.^16-19 Compared to absolute NIHSS score change, the NIHSS percent change offers a more accurate measurement by expressing change relative to the admission NIHSS score.²⁰

This study aims to explore the relationship between PVT and NIHSS percent change in AIS-LVO patients who have undergone successful reperfusion. We hypothesize that PVT+ is associated with a reduced likelihood of neurological improvement at discharge compared to PVT−.

Methods

CT perfusion-based assessment

Whole brain pretreatment CTP was performed on the Siemens Somatom Force (Erlangen, Germany) with the following parameters: 70 kVP, 200 effective mAs, rotation time 0.25 s, average acquisition time 60 s, collimation 48 × 1.2 mm, pitch value 0.7, 4D range 114 mm × 1.5 s. CTP images were then post-processed using RAPID commercial software (IschemaView, Menlo Park, California, USA) for generating quantitative relative cerebral blood flow (rCBF) volumes, Tmax volumes, and qualitative Tmax maps.

Based on qualitative Tmax maps, PVT was assessed within the posterior SSS at the level of the lateral ventricle occipital horns for more proximal venous drainage and at the torcula for deep venous drainage, by experienced board-certified neuroradiologist (VY, 10 years of working experience and DL, 4 years of working experience). PVT+ was defined as Tmax ≥10 s timing within the posterior SSS, torcula, or both areas, and PVT− was defined as lacking Tmax ≥10 s timing within the SSS and the torcula (Fig. 1).

Results

Discussion

In this study of successfully reperfused AIS-LVO patients, PVT+ was independently associated with a lower NIHSS percent change and a reduced likelihood of achieving more than 70% improvement in NIHSS scores. These findings provide supportive evidence that VO impairment, as represented by PVT+, is a key determinant of short-term neurological changes in AIS-LVO patients. To our knowledge, this is the first study to report associations between VO and NIHSS percent change.

The NIHSS is recognized as a reliable indicator of hospital length of stay, functional outcome at discharge, discharge disposition, and in-hospital death.^16-19 As it is routinely applied upon admission to measure neurological deficits, it is also considered a valuable tool for tracking neurological functional changes over time. However, the use of absolute NIHSS score change has a major limitation: a specific absolute change might indicate substantial improvement in patients with lower baseline NIHSS scores, while the same change might not significantly impact functional outcomes in patients with more severe strokes. The NIHSS percent change offers a more accurate measurement by expressing change relative to the admission NIHSS score.²⁰

Our study found that favorable VO was associated with lower admission NIHSS scores, consistent with previous research.²² Additionally, favorable VO was associated with a higher NIHSS percent improvement. These findings support the utility of PVT as a valuable adjunct imaging biomarker derived from CTP for assessing VO profiles in AIS-LVO. PVT has already demonstrated its utility in predicting 90-day functional outcomes^{14, 15} and this study further validates its association with short-term neurological changes.

Detection of PVT+ is notably more discernible on post-processed Tmax maps, aiding interpretation even for non-experts. The cortical vein opacification scores (COVES) scale is a 6-point system that requires evaluation of three specific veins (vein of Labbe, sphenoparietal sinus, and superficial middle cerebral vein).⁷ COVES has been served as a CTA imaging surrogate of VO assessment. A favorable VO has been associated with successful first-pass reperfusion in AIS-LVO,⁵ as well as with positive 90-day functional outcomes following MT for anterior circulation LVO.²³ Conversely, unfavorable VO correlates with a higher incidence of parenchymal hemorrhage,²⁴ early brain edema,² and poorer 90-day outcomes.²⁴ PVT is a binary parameter based on the assessment of two locations: the superior sagittal sinus (SSS) and the torcula. Compared to COVES, this approach simplifies the evaluation process, making it more practical for clinical settings. The superior sagittal sinus is indicative of superficial cerebral venous drainage while the torcula reflects deep venous drainage. Moreover, PVT+ is defined by the presence of prolonged perfusion time (Tmax ≥10 s), providing temporal information indicative of a severe delay relative to normal transit times. This dual aspect enhances its diagnostic value by capturing both static and dynamic aspects of venous circulation dysfunction in AIS-LVO. Importantly, the PVT status is determined solely by the presence of Tmax ≥10 s and is unrelated to the volume of Tmax >10 s (Fig. 1).

Our study adds to the growing body of literature highlighting the significance of VO in the microperfusion of ischemic brain tissue. We demonstrated that VO is associated with less neurological deficit improvement during hospitalization. Prior research by Heitkamp et al. suggested that unfavorable VO is a strong predictor of unfavorable functional outcomes despite successful recanalization.^{1, 9} Adusumilli et al. further illustrated that a favorable comprehensive venous profile is strongly associated with functional independence and excellent postthrombectomy reperfusion.¹¹ Additionally, Faizy et al. found that favorable VO correlates strongly with favorable tissue-level collaterals indicated by HIR <0.4, ultimately leading to better functional outcomes.^{25, 26} Our findings are in agreement with these observations, further corroborating the growing evidence demonstrating the influence of VO for prognostication.

Our study found that hemorrhagic transformation (HT) was associated with a reduced likelihood of achieving more than 70% improvement in the NIHSS score. HT is a common complication of acute ischemic stroke and is linked to poor clinical outcomes.²⁷ Previous research has shown that HT correlates with worse functional outcomes at discharge.²⁸ Specifically, parenchymal hematoma type 2 (PH2) has been associated with diminished neurological improvement following thrombolysis (Gill, 2016).²⁹ Unfavorable VO has been reported as a risk factor for hemorrhagic transformation, particularly parenchymal hematoma, after endovascular treatment.^{6, 24} In our cohort, while there was a higher incidence of PH2 (4.8% in PVT− vs. 13.9% in PVT+), the difference did not reach statistical significance, likely due to the low overall incidence of PH2 in our sample.

We acknowledge several limitations. Firstly, the retrospective design introduces inherent limitations such as selection bias and confounding factors. Additionally, the availability of CTP may be limited in smaller centers and rural locations, potentially impacting the generalizability of our findings. PVT is also subject to various factors, including the influence of blood flow, drainage patterns from non-acutely affected regions, and aspects of the imaging protocol such as the rate of contrast injection and bolus timing. Moreover, the PVT in our study was designed to assess only two representative locations for simplicity, potentially limiting its comprehensive representation and susceptibility to local pathologies like thrombosis. Given these limitations, future research should aim for multicenter, prospective studies to further investigate the strength of the association between VO profiles and stroke outcomes, as well as to assess the generalizability of our findings across diverse patient populations and settings.

Author Contributions

J.M, H.S, D.L, L.L, A.B, M.S, N.H, F.D, A.D, A.G, V.V, V.U, E.M, H.L, R.X, R.L, D.W, G.S, B.P, K.N, G.A, M.W, J.H, T.F, A.H, R.L, V.Y. contributed to the conception and design of the work. J.M, H.S, D.L, L.L, A.B, M.S, N.H, F.D, A.D, A.G, V.V, V.U, E.M, H.L, R.X, R.L, D.W, G.S, B.P, K.N, G.A, M.W, J.H, T.F, A.H, R.L, V.Y. were involved in the acquisition of data, and data analysis and interpretation. J.M, H.S, D.L, L.L, A.B, M.S, N.H, F.D, A.D, A.G, V.V, V.U, E.M, H.L, R.X, R.L, D.W, G.S, B.P, K.N, G.A, M.W, J.H, T.F, A.H, R.L, V.Y. drafted the work and revised it critically for important intellectual content. V.Y: Guarantor

Conflict of Interest

The authors have no conflict of interest to declare.

Funding Information

No funding information provided.