ECG-gated 23 Na-MRI of the human heart using a 3D-radial projection technique with ultra-short echo times

Abstract

Pathological changes in tissue often manifest themselves in an altered sodium gradient between intra- and extracellular space due to a malfunctioning Na+-K+ pump, resulting in an increase in total sodium concentration in ischaemic regions. Therefore, 23Na-MRI has the potential to non-invasively differentiate viable from non-viable tissue by detecting concentration changes of intra- and extracellular sodium. As the in vivo sodium signal shows a bi-exponential T2 decay, with a short component of less than 1 ms, the accurate quantification of the total sodium content requires imaging techniques with ultra-short echo times (TE) below 0.5 ms. A 3D-radial projection technique has been developed which allows the acquisition of ECG-triggered sodium images of the human heart with a TE of 0.4 ms. With this pulse sequence 23Na-MRI volunteer measurements of the head or the heart were performed in less than 18 min on a 1.5-T clinical scanner with an isotropic resolution of 10 mm3. The signal to noise ratio of the radial projection technique is twofold higher than that of a Cartesian gradient echo pulse sequence (TE = 3.2 ms). Radial 23Na-MRI provides a tool for clinical studies, aiming at the differentiation of viable and non-viable tissue.

Improved cardiac sodium MR imaging by density‐weighted phase‐encoding

“…The myocardial SNR, however, may not be a good criterion to assess sensitivity, as it can be influenced by varying degrees of contamination by blood signal from the ventricles. The SNR values reported for blood in the ventricles are 13.9 (2), 13.1 (15), and 17.6 (16), and 21.1 in our study. This difference in SNR can most likely be attributed to the different acquisition delays, which vary from 0.4 to 3.6 msec for the studies cited: with a T 2 * of about 13 msec (16), the acquisition delay has a strong influence on the SNR.…”

“…It is difficult to directly compare the SNR reported by various studies (2,15,16), as different field strengths, coils, and imaging parameters have been used. Renormalizing the literature values to an equivalent SNR definition, to a voxel volume of 1 mL and an experimental duration of 10 minutes, the SNR values reported for LV myocardium are 8.8 (2), 8.0 (15), and 11.7 (16), compared to 11.3 in our study.…”

Four-dimensional cardiac imaging in living embryos via postacquisition synchronization of nongated slice sequences

“…For larger organisms ͑from mice to humans͒, it is relatively easy to gate the acquisition with respect to electrocardiograms ͑ECG͒ or respiratory signals-a technique known as prospective gating or triggering-and reconstruct volumes at a fixed moment in the cycle. [5][6][7] Remaining motion artifacts may then be reduced by the use of various elastic registration procedures that warp the spatial data. [8][9][10] In cases where gating is not possible or unreliable, nongated dynamic datasets have been registered by a variety of methods and for various purposes.…”

Dynamic structure and protein expression of the live embryonic heart captured by 2-photon light sheet microscopy and retrospective registration

“…cardiac conduction or respiratory motion) to gate the 2D image acquisitions [4][5][6][7]. Alternatively, retrospective temporal registration of image stacks has been used to reconstruct the dynamic motion of cardiac tissues, taking advantage of the quasi-periodic nature of the heart motion [8].…”