Both contrast-enhanced (CE) and non-contrast enhanced MRA (NCE-MRA) methods have been widely used in clinical MR angiography (MRA) examinations. Since the introduction of abdominal CE-MRA in 1994 by Prince (1), the CE-MRA techniques have been advanced and applied to many area of body and peripheral run-offs. For estimating gadolinium (Gd) contrast arrival timing, test bolus and fluoroscopically triggering are used in clinical setting. In general, CE-MRA has lower spatial resolution due to fast Gd bolus chasing; however, time-resolved imaging of contrast kinetics (TRICKS) using view-sharing and periphery k space undersampling allows rapid and multiple image acquisition during fast passage of contrast bolus (2). Furthermore, Cartesian acquisition with projection-reconstruction-like sampling (CAPR) provides 1-mm isotropic resolution with 5-second frame rate for CE-MRA (3). In recent advancement of radial scan, Stack-of-Stars (radial VIBE) is reported to be motion insensitive in CE-MRA scan (4). The concerns of using Gd contrast agent rise because of recent reports on Gd deposition in the brain (5) as well as nephrogenic systemic fibrosis (NSF), which led to low dose Gd CE-MRA and re-introduction of ferumoxytol.

Due to safety concerns of Gd-based contrast material and reduction of cost, demand in development of NCE-MRA techniques has been considerably increased. In response, several new NCE-MRA methods have been developed, besides time-of-flight (TOF) and phase contrast (PC). These new techniques can be characterized by utilizing: inflow effect such as Quiescent-interval single-shot (QISS) (6); flow-dependency on cardiac phase such as fresh blood imaging (FBI) (7,8), and flow-sensitive dephasing (FSD) (9); and arterial spin labeling (ASL) techniques (10, 11) with flow-in, flow-out, and tag-on/off alternate acquisitions. The primary applications, advantages, and limitations of established and emerging NCE-MRA techniques are discussed.