Towards quantification of pulsed spinlock and CEST at clinical MR scanners: an analytical interleaved saturation–relaxation (ISAR) approach

Abstract

Off-resonant spinlock (SL) enables an NMR imaging technique that can detect dilute metabolites similar to chemical exchange saturation transfer. However, in clinical MR scanners, RF pulse widths are restricted due to recommended specific absorption rate limits. Therefore, trains of short RF pulses that provide effective saturation during the required irradiation period are commonly employed. Quantitative evaluation of spectra obtained by pulsed saturation schemes is harder to achieve, since the theory of continuous wave saturation cannot be applied directly. In this paper we demonstrate the general feasibility of quantifying proton exchange rates from data obtained in pulsed SL experiments on a clinical 3 T MR scanner. We also propose a theoretical treatment of pulsed SL in the presence of chemical exchange using an interleaved saturation-relaxation approach. We show that modeling magnetization transfer during the pauses between the RF pulses is crucial, especially in the case of exchange rates that are small with respect to the delay times. The dynamics is still governed by a monoexponential decay towards steady state, for which we give the effective rate constant. The derived analytical model agrees well with the full numerical simulation of the Bloch-McConnell equations for a broad range of values of the system parameters.

Towards an analytic solution for pulsed CEST

“…The second step in the derivation of the pulsed analytic solution is to connect the CW solution during t p,hard to the free recovery solution during t d,hard , which has the same form as the biexponential solution utilized in quantitative MT experiments . The steady‐state solution is found by equating the result before and after a pause‐pulse repetition, an approach employed in previous work . This calculation is outlined in the Appendix (Section A4).…”

“…The first two terms of R 1 ρ ,pul in Equation describe saturation effects (that match previous results), and the last term describes transferred solute rotation effects (and are new results). The saturation effects are a weighted average of what the rotating frame longitudinal relaxation rate would be under CW irradiation ( R 1 ρ ) and R 1a , with weighting factors corresponding to relaxation during and between pulses, respectively.…”

“…(In practice, such isolation of exchange contributions using typical AREX approaches is not possible because of direct water effects and contributions from confounding exchanging pools to the reference scan acquired at the frequency on the opposite side of the water resonance.) This labeling efficiency approach can be extended to pulsed saturation when the pulses are long, and no rotation effects are included . In this no‐rotation case, the labeling efficiency is multiplied by DC hard , as this factor scales the first term in R 1 ρ ,pul in Equation .…”

“…Recently, solutions that explicitly account for pulsed saturation have been derived, but with highly limited regimes, e.g. using repeated hard pulses or limited to very fast exchange . All of these methods ignore solute rotation and cannot account for simple observed effects, such as the optimum sensitivity when using a train of 180° pulses .…”

Quantitative pulsed CEST-MRI usingΩ-plots

“…To extend the Ω ‐plot method for the case of pulsed CEST, we used the solution for pulsed spin‐lock by Roeloffs et al . . Under the assumption of R 1ρ decay during the saturation pulse of duration t p for each individual pulse and R 1A recovery during the interpulse delay t d , the steady‐state Z ‐spectrum in the pulsed CEST experiment is given by where is the average longitudinal relaxation rate in the rotating frame.…”