The purpose of this study is to demonstrate the feasibility of simultaneous dynamic measurement of renal R2, R2’ and R2* by the means of a new method denoted psMASE-ME, in which a periodic π pulse shifting multi-echo asymmetric spin echo (psMASE) sequence is adopted together with a moving estimation (ME) strategy.

Blood oxygen level dependent (BOLD) magnetic resonance imaging (MRI) has been applied in evaluating intra-renal oxygenation status both in animal models and in humans. However, R2* is a summation of irreversible (R2) and reversible(R2’) relaxation rates, where R2 is sensitive to tissue water content and inflammation, and R2’ is able to provide more direct evaluation of renal oxygenation.

Recently, a multiple spin- and gradient-echo (SAGE) sequence with echo-planar imaging (EPI) acquisition scheme [1] was proposed which allows simultaneous quantification of R2 and R2* during a dynamic event. However, due to the non-ideal slice profile caused by short RF pulse duration, SAGE method suffers a problem of slice profile mismatch between images prior to and after the 180° refocusing pulse, which results in additional estimation errors. Other methods such as GESSE [2] and GESFIDE[3] could not achieve dynamic imaging. So we propose a new method denoted psMASE-ME for dynamic measurement of renal R2, R2’ and R2*.

psMASE Sequence: The diagram of the sequence is illustrated in Fig.1a. In the psMASE sequence, following each 90° excitation, the 180° refocusing pulse is shifted periodically. In each sub-sequence with a specific 180° pulse position, data are subsequently sampled by multi-EPI acquisition scheme with different offset from the spin echo. In this study, a four-echo acquisition scheme is adopted. Besides, to further improve temporal resolution of the imaging, a moving estimation strategy inspired from the signal-processing field [4] is introduced, which makes the re-use of the acquisition data at adjacent time points possible. The estimation principle of the psMASE-ME method is shown in Fig.1b. The moving window length is set as the period of the psMASE sequence, which is chosen to be 3TR in this study.

Phantom Study: In order to evaluate the accuracy of this psMASE-ME method, experiments were performed on a phantom consisted of 17 separate sample tubes containing 0.6, 0.7, 0.8, 0.9, 1.0, 1.1, 1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 2.0, 2.1 and 2.2 ml samples of gadopentetic acid meglumine (0.5 mmol/ml) and 200 ml normal saline (0.9% sodium chloride) to produce a range of R2 and R2* values. Imaging parameters for the psMASE sequence were: field of view (FOV) = 150 × 150 mm2, matrix size = 70 × 70, repetition time (TR) = 2000 ms, TE1/TE2/TE3/TE4 = 60/80/100/120 ms, echo space = 20 ms, τ = -10/0/10 ms, slice thickness = 6 mm, SENSE factor = 2. In addition, the psMASE-ME derived R2 and R2* were compared to reference values obtained in separate acquisitions using multi-echo spin echo (MSE) and multi-echo gradient echo
(MGE) sequences, respectively.

In Vivo Study: Eight rabbits (2.5-3.5 kg) were included for the study. Continuous psMASE imaging was performed during a respiratory challenge. A 4-min baseline data was first acquired while the rabbits breathed air. Then the gas regime was sequentially changed at 6-min intervals: carbogen (97% O₂, 3% CO₂) - air - 100% O₂ - air.

In phantom study, strong correlations were found between the proposed psMASE-ME derived R2/R2* and those obtained by MSE/MGE methods, with correlation coefficients (r) of 0.999 and 0.996 respectively. The psMASE-ME derived R2, R2’ and R2* maps of one typical rabbit acquired during respiratory challenge are represented in Fig. 3. The average time curves of R2, R2’ and R2* in cortex are displayed in Fig.4. In both cortex and medulla, R2, R2’ and R2* under carbogen and pure oxygen challenge decreased significantly compared with air breathing. No statistical difference was found between carbogen and oxygen challenge (P = 0.38).This study demonstrates the feasibility of dynamic renal R2, R2’ and R2* measurement with the proposed psMASE-ME method. High temporal resolution was achieved by combining psMASE sequence and moving estimation strategy. The superiorities of this method could potentially be used in the renal oxygenation evaluation.