FLAIR vascular hyperintensity combined with asymmetrical prominent veins in acute anterior circulation ischemic stroke: prediction of collateral circulation and clinical outcome

Acute ischemic stroke (AIS) has a large burden of morbidity and mortality worldwide, accounting for 87% of all total strokes [1]. Identification of irreversible infarct core and salvageable penumbra is essential to guide the therapeutic decision. The clinical practice found that there was a mismatch between the volume of infarction and the therapeutic effects, as well as the functional prognosis. This diversity may be partly explained by the different compensatory capacities of collateral circulation [2, 3]. Good collaterals enable timely reperfusion of blood flow to the ischemic penumbra, including reducing the volume of cerebral infarction, prolonging the therapeutic window, decreasing neurological impairments, and improving prognosis [4]. Recently, imaging techniques with contrast media have been introduced for the evaluation of collateral circulation, however, considering the complication of irreversible renal damage caused by contrast injection, and the inability to perform it in patients with renal insufficiency or contrast medium allergy, it is necessary to find a more non-invasive imaging method to estimate collaterals [5].

Fluid-attenuated inversion recovery (FLAIR) is now routinely used to improve the detection of lesions in patients with AIS because of its ability to suppress cerebrospinal fluid hyper-signal. FLAIR vascular hyperintensity (FVH) was probably due to compensatory reflux from the distal to the proximal of a stenotic or occluded vessel, as the slow blood flow would lead to a loss of the flowing void effect and result in hyperintensity, serving as a non-contrast visualization marker of leptomeningeal collaterals [6‐8]. When the primary collateral circulation cannot provide enough blood perfusion, the secondary and tertiary collateral vessels will be established. Among them, the opening of leptomeningeal collaterals is an important pathway. FVH could serve as an objective basis for the formation of leptomeningeal collateral circulation after stroke [9]. Many studies have confirmed that FVH was correlated with circulation as well as the severity of hemodynamic impairments, opening up a new idea of exploring noninvasive assessment of the status of intracranial collaterals and prognosis [10‐12].

Susceptibility-weighted imaging (SWI) is a novel, high-resolution, three-dimensional gradient-echo MR technique, characterized by being highly sensitive to paramagnetic substances and has been implemented in the assessment of cerebral hemodynamics following AIS [13]. Asymmetrical dilated-vessel-like signal loss seen in the ischemic cerebral hemisphere on SWI, namely asymmetrical prominent veins sign (APVS), can indirectly show an increase of oxygen extraction fraction (OEF) in the penumbral, has been increasingly recognized as an alternative to assess collaterals and an imaging biomarker for predicting prognosis [14‐16]. It has been widely believed that the APVS correlated with uncoupling between the oxygen supply and demand in the hypoperfusion tissue, resulting in a relative increase in deoxyhemoglobin [17]. To our knowledge, no studies have been found on exploring the correlation between the combination of FVH and APVS in the evaluation of collateral circulation in AIS, and predicting the functional prognosis. Therefore, this study aimed to investigate the relationship between FVH, APVS and prognosis, to explore the correlation between the status of collaterals and relevant factors affecting prognosis, and to provide an imaging basis for the blood perfusion status in the ischemic area, individualized precise treatment and prognosis evaluation.

Our results showed a positive correlation between the FVH-ASPECTS and APVS-ASPECTS, suggesting that the presence of extensive FVH and APVS in the acute phase of stroke can both reflect the status of collaterals. We also found that the significant predictors of a poor prognosis in AIS treated without recanalization due to occlusion or stenosis of the unilateral ICA and/or M1 segment of the MCA within 72 h of symptom onset were higher baseline NIHSS scores, larger baseline DWI lesion volume, more extensive FVH and APVS grading. In addition, the combination of the FVH and APVS was more predictive of poor prognosis than either alone.

Good collateral flow is increasingly recognized as a crucial element in therapeutic and prognostic assessment, as it can markedly reduce penumbra loss and prolong the salvaged time [27]. The new concept of collateral histology reveals that collateral circulation includes the entire cerebral circulatory system including the arterial, microvascular and venous systems. Consequently, collateral vessels can be judged indirectly via capillary horizontal perfusion, metabolic brain tissue status and changes in intracranial drainage venous pressure besides direct visual imaging technology [28]. FVH and APVS can reflect the hypoperfusion state of damaged brain tissue in patients with AIS and provide information on blood perfusion for assessing the status of the collateral circulation. This study used the ASPECTS score system to quantify the extent of FVH and APVS, and analyzed the correlation between them on this basis. Regarding FVH and APVS, previous studies reached a consensus about their mechanism that patients with severe stenosis or occlusion of large vessels were more likely to have extensive FVH and APVS [15, 29]. When the infarction caused by severe stenosis or occlusion of large vessels occurred, brain tissue in the infarct core area experienced a sequence of pathophysiological mechanisms, including ischemia, hypoxia, cell edema and so on. Under the stimulation of ischemia and hypoxia, the establishment and opening of the leptomeningeal collaterals occurred, along with changes in brain oxygen metabolism and vein expansion, resulting in an increase in oxygen uptake fraction to maintain the basic metabolic level of brain cells. Our study also demonstrated a positive correlation between FVH score and APVS score, which further confirmed the hemodynamic mechanisms of FVH and APVS from both arterial and venous perspectives. Despite the distinct physiological mechanisms of FVH and APVS, they both can indicate inadequate perfusion of brain tissue and reflect the status of collateral circulation in the acute phase.

In clinical practice, research has shown a higher incidence of disability in individuals with anterior cerebral circulation stroke. Subsequently, the early identification of useful prognostic markers plays a critical role in treatment selection and timely intervention. Our study discovered that patients with poor outcomes exhibited increased occurrences of atrial fibrillation in their medical history, higher NHISS scores at baseline, larger baseline infarct volumes, and more extensive FVH and APVS. We suggested that a history of atrial fibrillation is associated with a poor prognosis, possibly due to the abrupt onset of stroke in patients with cardiogenic embolism, the lack of time for collateral branches to form around the ischemic tissue and insufficient compensation of collaterals can result in more severe hypoperfusion [30]. This also highlights the significance of distinguishing stroke subtypes caused by different pathological mechanisms for predicting functional prognosis. The NIHHS score is a widely-used scale for assessing the severity of neurological deficits and plays a crucial role in predicting their functional prognosis. Generally, higher NIHHS scores indicate greater neurological impairment and poorer functional outcomes, and vice versa. A further logistic regression analysis showed that the NIHSS score was an independent risk factor for a poor prognosis. The volume of infarction strongly predicted clinical prognosis has been reported earlier [31]. Our study similarly identified an independent correlation between infarct volume and poor prognosis in AIS patients. The larger the irreversible damage volume of brain tissue, the more severe the clinical dysfunction and unfavorable recovery of the ischemic semidark zone, and the worse the prognosis.

Currently, research on the relationship between FVH and functional outcomes varies with time. The duration of time from symptom onset to imaging has been reported as an essential factor for evaluating the prognostic value of FVH [32]. In most studies that suggested a correlation between FVH and favorable outcomes, the time from symptom onset to imaging was less than 6 h [33‐35]. In comparison, in the majority of studies that demonstrated an association between FVH and unfavorable outcomes, the symptom-to-imaging time was 12–24 h or longer [36‐38]. Our findings confirmed that FVH was independently associated with poor prognosis, which we believe is related to the fact that the patients we enrolled who did not receive timely vessel recanalization and underwent MR scanning within 72 h. Previous studies believed that FVH may serve as an imaging marker of leptomeningeal collaterals within 6 h of symptom onset or within the time window of reperfusion treatment [39]. The extent of FVH may indicate the amount of brain tissue vulnerable to ischemia, which can be reversed with reperfusion therapy to reduce the size of the eventual lesion and improve functional outcomes. This clarifies why FVH-positive patients tend to have superior clinical outcomes compared to those FVH-negative patients during this specific time interval. However, when FVH appears beyond the time window for reperfusion therapy, it may indicate persistent occlusion of the vessel and hypoperfusion. Thus, patients with FVH may be more susceptible to hemodynamic instability than patients with and without FVH. This variability might be related to the severity of clinical impairment and unfavorable outcomes in patients with FVH during this period. We recommended further investigation into the potential mechanisms underlying this difference.

APVS was found to be another independent factor associated with a 90-day poor prognosis in our study. To some extent, extensive APVS represents an acute ischemic penumbra with poor perfusion. As the hypoperfusion time increases, the size of the ischemic penumbra diminishes, which is predictive of early neurological deterioration (END) and stroke progression [16]. Several studies have investigated the correlation between APVS and prognosis. Some argued that APVS was not associated with prognosis [40, 41], while others believed that patients with APVS had a worse prognosis than those patients without APVS [26, 42, 43]. Chen et al. [40] considered that extensive APVS was associated with poor prognosis, whereas patients without APVS tended to have better collaterals compensation and no obvious hypoperfusion area, thus making it easier to obtain a good prognosis. Sun et al. [44] demonstrated that APVS was a predictor of poor prognosis and was independently associated with END. In our study, patients with extensive APVS had a worse functional prognosis, possibly due to the absence of thrombolytic treatment. Conservative therapy was primarily given to individuals with contraindications to thrombolytic or over a time-window, with more underlying conditions. Although collateral circulation existed at this time, it was more susceptible to stroke progression due to the inability to receive timely and effectively. As a result, it carries a higher overall probability of resulting in a poor prognosis [45]. This outcome may provide clinicians with novel insights into the precise management of stroke, as timely endovascular therapy within the time window is crucial for improving prognosis. We further clarified the value and efficacy of FVH and APVS alone and their combination in predicting poor outcome by ROC curves. The combination AUC was higher than that of each individual index and can compensate for the limitations of the evaluation of each individual index. These findings also offered a degree of reference for the clinical treatment assessment and functional prognosis prediction.

There are several limitations in this study. It is a retrospective analysis and has a potential risk of selection bias, without analysis of patients who underwent thrombolysis or endovascular therapy. Infarct volume determined by ASPECTS is influenced by the quality of the images and this may lead to bias; Finally, FVH and APVS were not observed dynamically in this study, a longitudinal study investigating the association with the change of FVH and APVS and the progression of collateral as well as prognosis is needed.

When stroke occurred, apart from the rapid establishment and opening of the leptomeningeal collaterals in the ischemia and hypoxia area, there is also an increase in the oxygen uptake fraction of brain tissue to co-compensate for the ischemic and hypoxic state. The presence of FVH and APVS can comprehensively assess collateral circulation from both the vein and artery perspective, and there is a positive correlation between the two. Besides, the two indicators were independently related to a poor prognosis at 90 days, and their combination can enhance the predictive accuracy. Using the FLAIR and SWI sequence together is useful for an early non-invasive assessment of collateral status, the selection of individualized treatment and improves the prediction of clinical outcome. Further studies with a larger sample size are needed to confirm our results.