• Xin Li, Charles S Springer, and Michael Jerosch-Herold.
• Advanced Imaging Research Center, Oregon Health & Science University, Portland, OR 97239, USA. lxin@ohsu.edu
• NMR Biomed. 2009 Feb 1; 22 (2): 148-57.

AbstractHuman myocardial (1)H(2)O T(1)-weighted dynamic contrast-enhanced MRI data were acquired during the brief first-pass period after injection of a very small gadolinium diethylenetriaminepenta-acetate (GdDTPA(2-)) dose. The shutter-speed pharmacokinetic effects of both transendothelial and transcytolemmal equilibrium water exchange processes were investigated. Our results indicate that even for such a short acquisition window and relatively large pseudo-first-order rate constant (K(trans)) for plasma/interstitium contrast reagent (CR) transfer the kinetics of these water exchange processes cannot be treated as infinitely fast or slow. However, neither the intracellular water molecule lifetime (tau(i)) nor its intravascular counterpart (tau(b)) are among the parameters most influential in analysis of the noisy data typically associated with the cardiac perfusion application. Thus, the actual values of water exchange kinetic rate constants are relatively indeterminate as this experiment is usually conducted. Combining the K(trans) evaluations with independently determined flow (F) values allows us to estimate CR permeability coefficient surface area product (P(CR)S) values. The fact that the P(CR)S magnitudes almost equal the K(trans) values confirms that GdDTPA(2-) extravasation in resting human myocardial muscle is indeed permeation-limited and supports the validity of the K(trans) and P(CR)S estimations. Nevertheless the model analysis is most consistent with the results if P(CR)S is not assumed to be constant with changing flow. The capillary blood volume fraction (v(b)) is a sensitive parameter in the analysis. We also compared resting and hyperemic cardiac conditions, the latter resulting from the volume flow increase induced by adenosine arteriolar vasodilation. We found that the P(CR)S value increases with flow probably mostly because of an S increase associated with capillary recruitment. The v(b) values also increased in hyperemia and showed a flow-dependence with a clearly identifiable component due to capillary recruitment.Copyright (c) 2008 John Wiley & Sons, Ltd.

26 keywords...

hide…