To assess the reliability and reproducibility of pseudo‐continuous arterial spin labeling (pCASL). Intra‐ and interscanner reliability and reproducibility are essential before pooling perfusion imaging data.

Eight volunteers were scanned three times on two 3.0T magnetic resonance imaging (MRI) scanners using the pCASL technique. The first and third tests were conducted on scanner one, while the second test was on scanner two. The pCASL was acquired with postlabeling delay time (PLD) of 1.5 and 2.5 seconds. The cerebral blood flow (CBF) of gray matter, white matter, frontal lobe, parietal lobe, occipital lobe, temporal lobe, putamen, posterior cingulate, and thalamus were extracted for comparison. Reliability and reproducibility were measured using the intraclass correlation coefficient (ICC) and within‐subject coefficients of variation (wsCV), respectively.

The pCASL consistently had high intra‐ and interscanner measurement reliability and reproducibility. Reproducibility of the CBF values was higher, with PLD 2.5 seconds than with PLD 1.5 seconds. Different PLDs had considerable influence on the reproducibility of the different regions of the brain.

Although excellent intra‐ and interscanner reliability and reproducibility could support the feasibility of cross‐site pooling of ASL data, pCASL with multiple PLDs may better assess the CBF of the human brain. J. Magn. Reson. Imaging 2014;39:402–409. © 2013 Wiley Periodicals, Inc.