Detection and evaluation of hemodynamic signal changes during reperfusion after cerebral ischemia play an important role in the clinic to estimate the treatment outcome. Multi-parametric MR images such as perfusion, diffusion, and water relaxation times (T₁ and T₂) have been used to detect the signal changes after ischemia. In recent years, amide proton transfer (APT) imaging has become an important method as a biomarker of pH in ischemia due to the high sensitivity of the tissue acidosis, which depends on the changes of amide proton exchange rate.^1-4 However, unlike permanent ischemia, the application of APT imaging to transient ischemia have not been widely studied yet. The purpose of this study is to evaluate the ischemic tissue injury and the response to the reperfusion in rats using APT metrics (conventional MTR_asym(3.5ppm) and pure APT^#/NOE^# signals based on the extrapolated semi-solid magnetization transfer reference signals (EMR)) and multi-parametric MR images.⁵

MRI experiment: Transient middle cerebral artery occlusion (MCAO) was induced in four male Wistar rats by inserting a nylon suture filament into the lumen of the internal carotid artery to block the MCA, and remained for 3 hours. The reperfusion process was achieved by removal of the filament. The MR imaging was obtained using a 4.7T Bruker scanner at different time points (before reperfusion, immediately after reperfusion, 1 day, 3 days, and 7 days after reperfusion). The CEST datasets with 61 frequency offsets (S₀ image and -15~+15ppm at intervals of 0.5ppm) were acquired using a long continuous-wave RF saturation pulse (power = 1.3 μT, saturation time = 4 sec). For B₀ corrections, WASSR datasets with 26 frequency offsets (-0.6~+0.6ppm at intervals of 0.05ppm) were acquired using 0.5 μT RF saturation power. In addition, high SNR APT images were acquired using two frequency offsets (±3.5ppm) with sixteen signal averages. For multi-parametric MR images, the ADC map with seven b-values (0~1000 s/mm²), the CBF map with arterial spin labeling (ASL) technique, the T_1w map with seven IRs (0.05~3.5 s), and the T_2w map with seven TEs (30~90 ms) were acquired.

Data processing: The MTR_asym(3.5ppm) image was calculated using B₀ corrected datasets with the following equations: S_sat(-3.5ppm)/S₀ - S_sat(+3.5ppm)/S₀. For the APT^# and NOE^# images, the B₀ corrected datasets were fitted to Henkelman's two-pool MTC model with the super-Lorentzian lineshape.⁶ To avoid possible CEST and NOE contributions, only limited data points (+15~+7ppm) were fitted. The EMR signals (Z_EMR) in the whole offset frequency ranges were obtained using acquired MTC model parameters, and the differences between Z_EMR and experimental data at 3.5ppm (APT^#) and -3.5ppm (NOE^#) were calculated. The ADC, T_1w, T_2w maps were fitted with the following equations: I=I₀∙exp(-b∙ADC), I=I₀+A∙exp(-TI/T_1w), and I=I₀∙exp(-TE/T_2w). The CBF map was reconstructed from images with and without labeling.

Before the reperfusion process, the decreased APT^# signals were clearly observed in the ischemic lesions, and the observed NOE^# signals were larger than the APT^# signals (Fig. 1). Due to this, the observed MTR_asym(3.5ppm) signals in ischemic lesion became quite negative. After the reperfusion process was performed, as expected, the APT^#, NOE^#, and MTR_asym(3.5ppm) signal differences between contralateral and ischemic lesions were getting smaller until reperfusion on 3 days. Notably, the edema was appeared in the ischemic legion as the recovery process, and it may affect the signal changes in the APT^#, NOE^#, and MTR_asym(3.5ppm).

The ADC, APT^#, and MTR_asym(3.5ppm) signals have significant differences in the before and after reperfusion stages (p < 0.05) (Fig. 2). Between 1 day and 3 days, the ADC, CBF, APT^#, NOE^#, and MTR_asym(3.5ppm) signals were no significant differences. Notably, on 7 days after reperfusion, the NOE^# signals between contralateral and ischemic lesion were significantly changed, and these may affect the MTR_asym(3.5ppm) signal changes. The T_1w and T_2w values in the ischemia lesion were also significantly changed after reperfusion process.

The reconstructed images also clearly showed the signal changes of ischemic lesion before and after reperfusion (Fig. 3). The APT^# and MTR_asym(3.5ppm) signals were generally lower in the ischemic lesion than the contralateral; however, the lower signals in the ischemic lesion were recovered after the reperfusion process.

The application of APT imaging to ischemia and its reperfusion process showed the significant image contrasts between contralateral and ischemic lesion, which are highly related with the changes of the tissue acidosis. The overall results of the temporal evolution of the signal changes from the APT metric and the multi-parametric images were clearly indicated that APT imaging can also be a useful technique to estimate the infarction status in the ischemia reperfusion stages.