Abstract

Arterial spin labeling (ASL) is a non-contrast method of measuring cerebral perfusion with MRI. It has several advantages over traditional contrast-based perfusion-weighted imaging, including non-invasiveness, more straightforward cerebral blood flow (CBF) quantification, and repeatability. However, because of its lower signal-to-noise ratio (SNR) per unit time and its high sensitivity to arterial transit delays, it has not been used frequently in acute stroke, where arterial delays and time-efficiency are of the essence. This is beginning to change, driven by higher SNR implementations of ASL and the increasing use of 3T scanners. Furthermore, velocity-selective ASL sequences that are largely insensitive to arrival times are beginning to be applied to patients with cerebrovascular disease and promise the ability to quantify CBF even in regions supplied by late-arriving collateral flow. Despite these advances, many practical issues must be addressed to optimize ASL for its use in acute stroke studies. These include optimizing the trade-off between time, SNR, imaging resolution, and sensitivity to slow flow. Rapid and robust post-processing of image data must be made routine, such that CBF maps are available in real time so that they can be considered when making treatment decisions. Lastly, automated software needs to be developed in order to delineate hypoperfused tissue volumes, which is challenging due to the inherent differences between gray and white matter CBF. Attention to these details is critical to translate this promising research tool into mainstream clinical trials and practice in acute stroke.

View details for DOI 10.1007/s12975-012-0159-8

View details for PubMedID 24323778

View details for PubMedCentralID PMC4040499