T2* measurement during first-pass contrast-enhanced cardiac perfusion imaging

Laboratory of Cardiac Energetics, National Heart, Lung and Blood Institute, National Institutes of Health, DHHS, Bethesda, Maryland 20892-1061, USA.
Magnetic Resonance in Medicine (Impact Factor: 3.57). 11/2006; 56(5):1132-4. DOI: 10.1002/mrm.21061
Source: PubMed


First-pass contrast-enhanced (CE) myocardial perfusion imaging will experience T(2) (*) effects at peak concentrations of contrast agent. A reduction in the signal intensity of left ventricular (LV) blood due to T(2) (*) losses may effect estimates of the arterial input function (AIF) used for quantitative perfusion measurement. Imaging artifacts may also result from T(2) (*) losses as well as off-resonance due to the bolus susceptibility. We hypothesized that T(2) (*) losses would not be significant for measurement of the AIF in full-dose studies using a short echo time (TE = 0.6 ms). The purpose of this study was to directly measure T(2) (*) in the LV cavity during first-pass perfusion. For single-dose Gd-DTPA (0.1 mmol/kg at 5 ml/s), the LV blood pool T(2) (*) had a mean value of 9 ms (N = 10) at peak enhancement. Distortion of the AIF due to T(2) (*) signal intensity loss will be less than 10% using TE = 0.6 ms.

Full-text preview

Available from:
  • Source
    • "The effects of non-uniform magnitude modulation during data acquisition were also numerically simulated for three pairs of (T1, T2*) values in milliseconds: (120, 20), (13, 5), and (5, 2). The T2* values were estimated as approximately half of the T2, based on first-pass measurements in the left ventricle [12]. The T1 and T2 relaxation times correspond to CA concentrations in the RV of ≈1.6, 16 and 46 mmol/L, using relaxivities r1=4.5, r2=5.1 L mmol−1 sec−1. "
    [Show abstract] [Hide abstract]
    ABSTRACT: First-pass myocardial perfusion is often imaged with a tailored hybrid centric interleaved echo-planar-imaging sequence, providing rapid image acquisition with good contrast enhancement. The centric interleaved phase-encode order minimises the effective time-of-echo but it is sensitive to frequency-offsets. This short article aims to show possible artefacts that might originate with this sequence, in the context of first-pass perfusion imaging, when frequency-offsets are present. Non-uniform magnitude modulation effects were also analysed. Numerical and phantom simulations were used to illustrate the effects of frequency-offsets and non-uniform magnitude modulation with this sequence in a typical perfusion protocol. In vivo data was post-processed to analyse the h-EPI's sensitivity to the frequency-offsets. The centric phase-order was shown to be highly sensitive to frequency-offsets due to its symmetrical phase slope. Resulting artefacts include blurring, and splitting of the image into two identical copies along the phase-encode direction. It was also shown that frequency-offsets can introduce signal loss and ghosting of the right ventricle signal into the myocardium. The in vivo results were confirmed by numerical and phantom simulations. Magnitude modulation effects were found to be small. Imaging first-pass myocardial perfusion with an hybrid centric echo-planar-imaging sequence can be corrupted with ghosting and splitting of the image due to frequency-offsets.
    Journal of Cardiovascular Magnetic Resonance 06/2012; 14(1):44. DOI:10.1186/1532-429X-14-44 · 4.56 Impact Factor
  • Source
    • "For example, measurements of regional myocardial blood flow (in ml/g/min) can be made using time-intensity curve deconvolution with a measured arterial input function (AIF), usually from the left ventricular cavity or ascending aorta[29,30]. Measurement of the true AIF is affected by short T1s and T2* effects[31] that occur with high contrast agent concentrations during bolus contrast agent passage in the blood pool. This has led to two imaging strategies: 1) the "dual-bolus" method[32,33], which involves AIF measurement during a separate low-dose injection (1/10th or 1/20th the dose) prior to the full-dose injection used for the measurement of myocardial enhancement and 2) the "dual-sequence" method [34-36], which replaces one of the imaging slices with an AIF measurement slice that employs an acquisition that avoids the saturation and T2* effects at high concentrations. "
    [Show abstract] [Hide abstract]
    ABSTRACT: In less than two decades, first-pass perfusion cardiovascular magnetic resonance (CMR) has undergone a wide range of changes with the development and availability of improved hardware, software, and contrast agents, in concert with a better understanding of the mechanisms of contrast enhancement. The following review provides a perspective of the historical development of first-pass CMR, the developments in pulse sequence design and contrast agents, the relevant animal models used in early preclinical studies, the mechanism of artifacts, the differences between 1.5T and 3T scanning, and the relevant clinical applications and protocols. This comprehensive overview includes a summary of the past clinical performance of first-pass perfusion CMR and current clinical applications using state-of-the-art methodologies.
    Journal of Cardiovascular Magnetic Resonance 02/2008; 10(1):18. DOI:10.1186/1532-429X-10-18 · 4.56 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Concerns the output trajectory tracking for a class of nonlinear nonminimum-phase systems, namely, systems described by Lagrange equations of motion. In the nonminimum-phase case, the direct application of classical input-output inversion procedure fails, due to the unacceptable nature of internal dynamics. The paper concerns the inversion of those internal dynamics. Inversion yielding a bounded state permits then to achieve exact output tracking (or asymptotic tracking), without control saturation. The inversion scheme deals equally with systems having hyperbolic or nonhyperbolic equilibrium points of zero dynamic. This is a consequence of a different formulation of the stable inversion problem. The nonlinear internal dynamics are inverted, without any noncausal computation and are fully based on nonlinear terms, without local dynamic linearization. The performance is demonstrated through application to the tip trajectory tracking for a two-link flexible manipulator
    Decision and Control, 2001. Proceedings of the 40th IEEE Conference on; 02/2001
Show more