Introduction

Pulse train saturation has been used as an alternative for continuous wave in Chemical Exchange Saturation Transfer (CEST) MRI with reduced power deposition as well as demand on hardware. Additionally, pulse trains have been utilized in novel ways to frequency label exchange¹ and even filter exchange rates^2,3. These methods are related to chemical exchange kinetics which means they are sensitive to both labile proton concentration as well as exchange rate which can be modulated by pH. However, there are some applications in which it may be advantageous to assay concentration independent to changes in pH such as phosphocreatine and creatine modulation in metabolically impaired muscles. In this study, we demonstrate a pulse scheme that uses a low duty cycle pulse train of π-pulses to achieve saturation that is dependent on labile proton concentration yet relatively insensitive to exchange rate.

Methods

Simulations were performed to explore the exchange rate sensitivity of π-pulsed and continuous wave CEST signals. Two sets of seven creatine phantoms were prepared: the first set at 30 mM in 1× phosphate buffered saline (PBS) and titrated to pH 6.36, 6.70, 6.97, 7.33, 7.65, 7.88, and 8.21; the second set at four different concentrations with different combination of four different pHs, i.e, 15 mM at pH 6.7, 7.0, 7.3, and 7.6, 30 mM at pH 7.0, 45 mM at pH 7.3, and 60 mM at pH 6.4 and imaged at room temperature. MR experiments were performed on a Bruker Biospec 9.4T/30-cm and a volume coil excitation and reception (4.0-cm ID). The chemical exchange-sensitive MR pulse sequence consists of a saturation preparation module for chemical exchange contrast and image readout. Images were read-out by a spin-echo EPI readout scheme: matrix size = 64x64, field of view = 50 x 50 mm, slice thickness = 5 mm, TR = 14 s and TE = 20 ms. Duty cycle dependence was studied using a train of π pulses with an average B₁ of 0.48 μT and 0.97 μT with different duty cycles (5%, 10%, 25%, and 50%).

Discussion

Concentration dependence of a π-pulsed saturation scheme was simulated with a fixed duty cycle of 10% and an average B₁ of 0.48 μT (20 Hz) in Fig. 1a. Each of these curves exhibits an exchange rate insensitive regime around the same exchange rate range and increase monotonically with increasing labile proton fraction. Therefore, these low duty cycle π-pulse saturation schemes maintain their concentration dependence and are useful for interrogating concentration of a labile fraction with limited pH effect interference within range of exchange rates that are insensitive. MTR asymmetry for a 30 mM creatine phantom at varying pH is shown in Figure 1b. Lower duty cycles exhibit ranges of rate insensitivity which in turn exhibit a signal that is relatively insensitive to pH. Creatine phantoms of increasing concentration and various pH levels exhibited variation in the asymmetry of their exchange related relaxation (R_ex,asym) under continuous wave saturation at 0.97 μT (Fig. 1c). While this relaxation did exhibit concentration dependence, higher pHs also increased R_ex,asym detracting from linear dependence on concentration (R̅² = 0.593). In contrast, when average R_ex,asym was calculated in the same phantoms using the average B₁ under π pulse saturation using a DC of 10% (Fig. 1d), there was a clear linear dependence (R̅² = 0.976) on concentration despite the disparity in pH.

Conclusion

Low duty cycle π-pulsed CEST is an easy to implement method for exchange rate insensitive CE-MRI measurement. It provides a labile concentration dependent signal that may be useful in studies in which a change of chemical exchange rate (e.g., pH, catalyst concentration, and temperature changes) can interfere with accurate assessment of pathology.

Acknowledgements

No acknowledgement found.

References

1. Yadav N, et al. Magn Reson Med 2013;69:966-973. 2. Zu Z, et al. Magn Reson Med 2013;69:637-647. 3. Xu J, et al. Magn Reson Med 2014;71:1798-1812.