Bolus delay and dispersion in perfusion MRI: implications for tissue predictor models in stroke.

Perfusion maps, which are calculated from dynamic-susceptibility contrast (DSC)-MRI data by deconvolution of the arterial input function (AIF), are commonly used to predict tissue infarction in acute stroke. However, since the AIF is commonly measured in a major artery, there can be perfusion measurement errors associated with bolus delay/dispersion. Although methods to account for delay-related errors have been proposed, the effect of dispersion is more difficult to deal with and is usually left uncorrected. This study presents an assessment of the delay-dispersion relationship in a group of patients. Although a significant correlation was observed with one of the bolus delay definitions used, the estimation of the dispersion from the degree of delay was unreliable. Importantly, the dispersion observed in many patients was sufficient to result in substantial perfusion errors. The results are compared with previous numerical simulations, and their implications for the assessment and management of acute stroke are discussed. Copyright (c) 2006 Wiley-Liss, Inc.