Acute ischemic stroke (AIS) is a debilitating disease, ranking 1st among causes of serious long-term disability and 5th among the causes of mortality [1]. Ischemia causes significant disruption to the blood-brain barrier (BBB), which predisposes patients to hemorrhagic conversion over days to weeks after the initial ischemic episode [2]. Symptomatic hemorrhagic conversion is a major stroke-related complication that occurs in 2–17% of patients with ischemic stroke and can significantly worsen patient recovery [[3], [4], [5], [6], [7], [8], [9], [10]]. However, determining the extent of BBB disruption remains challenging and is not a part of routine clinical care.

BBB disruption results in increased BBB permeability (BBBP), which fluctuates significantly during patient recovery. A neuroinflammatory response occurs within the first few days after ischemia, which increases the permeability further, causing a secondary injury [[11], [12], [13]]. The tight junctions of the BBB continue to degrade further during the acute phase of recovery, further increasing the risk of reperfusion injury and vasogenic edema. In the subacute and chronic stages of recovery from ischemia, neoangiogenesis and overexpression of tight junction proteins start the repair of the BBB, restoring its permeability [2]. Even though the extent of BBB disruption varies drastically depending on the ischemic insult and the timing from the initial injury, BBBP is yet to be taken into consideration in clinical decision-making. The two imaging modalities used historically to image BBBP are positron emission tomography (PET) and dynamic contrast-enhanced (DCE) MRI. The former modality is expensive, often not available in the inpatient setting, and requires a radioactive tracer, while the latter requires the use of gadolinium, which is often contraindicated in patients with AIS due to poor renal function, and is less sensitive at depicting regions of hyperemia in patients with AIS [14]. There is thus a pressing need to identify better imaging modalities that enable providers to routinely evaluate the BBBP on clinical scanners.

Arterial spin labeling (ASL) is non-contrast MRI sequence that has been well studied in patients with AIS and validated against DCE for quantifying cerebral blood flow (CBF) without the need for gadolinium [[14], [15], [16]]. Diffusion-prepared pseudocontinuous ASL (DP-pCASL) is a more recent ASL variant sequence that magnetically labels water molecules and tracks their movement across the blood-brain barrier [17], generating the water exchange rate (kw), which serves as a surrogate for BBB function and permeability [14,[17], [18], [19], [20]]. Water exchange across the BBB is mediated by a number of processes, including passive diffusion, active co-transport through the endothelial membrane, and predominantly by facilitated diffusion through the dedicated water channel aquaporin-4 (AQP4). AQP4 is located on the astrocytic end foot that interfaces with the endothelial cells of the capillaries, forming the BBB complex. AQP4 is upregulated during stroke and becomes critical for water homeostasis in the CNS by inducing vasodilation or vasoconstriction to adjust CBF as needed [21,22]. Early studies have validated DP-pCASL in AIS animal models [23,24] and in humans for other CNS pathologies [20,25,26], demonstrating reduction in kw in pathological states. Additionally, DP-pCASL has also been studied in healthy individuals, showing that age and gender significantly affect kw [27] However, it has not been studied in humans with AIS and it remains unclear why kw is reduced if AQP4 is upregulated and the BBB is more permeable in AIS.

Further, in order to corroborate DP-pCASL in stroke, we utilize a well-studied sequence that quantifies diffusivity in the extracellular and intracellular compartments of brain tissue, called Neurite Orientation Dispersion and Density Imaging (NODDI) [[28], [29], [30]]. DP-pCASL is used to calculate BBBP from a perfusion standpoint, while NODDI is used to characterize the ischemic burden on tissue microstructure and the resulting cytotoxic and vasogenic edema.

In our study, we used DP-pCASL and NODDI to quantify BBB integrity in patients with ischemic strokes. We hypothesize that the water exchange rate kw as derived by DP-pCASL will be significantly altered within the infarcted tissue and will vary depending on the time interval between symptom onset and imaging. We believe that this work will set the foundation for the routine assessment of BBB in medical care and clinical research using non-contrast neuroimaging techniques in an effort to minimize hemorrhagic conversion and optimize functional recovery.