Routinely used magnetization transfer (MT) asymmetry in measurement of pH-sensitive amide proton transfer (APT) effect is susceptible to concomitant contributions, including semisolid MT and nuclear overhauser effect (NOE). In this study, multi-pool contribution from NOE, MT, CEST at 2 ppm and APT during the acute stroke was resolved with a sum of five Lorentzian functions. Results confirmed the changes in MT and NOE offset each other, and reduction of APT was approximately two-fold of the CEST effect decrease at 2 ppm, dominating the commonly observed pH-sensitive MTRasym change during acute stroke.
Materials and methods
Adult male Wistar rats (n=8) underwent MRI 60 min after unilateral middle cerebral artery occlusion on a 4.7 T scanner. Multi-slice echo planar imaging (5 slices, slice thickness/gap=1.8/0.2 mm, FOV=20×20 mm2, matrix size=48×48) was acquired, including diffusion MRI ( b=250 and 1000 s/mm2, TR/TE=3250/42 ms, 16 averages, scan time=1.5 min), arterial spin labeling MRI (TR/TE=5400/20 ms, time of saturation (TS)=3400 ms, 32 averages, B1=4.7 μT, labeling distance of 15 mm, and scan time=2 min) and APT MRI (frequency offsets between ±6 ppm with intervals of 0.25 ppm, B1=1.5 μT, TR/ TS/TE = 6500/3000/10 ms). In addition, WASSR map was collected with B1=0.5 μT (frequency offsets between ±0.5 ppm with intervals of 0.05 ppm) for field inhomogeneity correction. Perfusion and diffusion lesions were manually segmented. Z-spectrum was flipped as 1- Sz/S0 and fitted with a sum of five Lorentzian functions corresponding to NOE, MT, direct water saturation, CEST and APT effects at -3.2, -1.5, 0, 2.0 and 3.5 ppm, respectively 4, and multi-pool contribution was quantified in ADC lesion, CBF lesion and their mismatch, and their corresponding contralateral normal regions. One-way ANOVA with Bonferroni correction was conducted and P<0.05 was considered statistically significant.Results
Fig. 1 shows manually segmented perfusion and diffusion lesions. Flipped Z-spectra from lesion and respective control areas were fitted with a sum of five Lorentzian functions, from which individual CEST effects of the five contributors to Z-spectra were resolved (Fig. 2). Figure 3 shows maps of the conventional MTRasym and resolved NOE, MT, CEST at 2 ppm and APT effect maps. In all the three lesion regions, MT effect increased significantly while CEST effects of NOE, CEST at 2 ppm, APT and routinely measured MTRasym decreased significantly (Fig. 4). Absolute change of CEST effects between ischemia and control regions were listed in Table 1. The change in MT and NOE has no significant difference among the three lesion regions, and furthermore, their comparably opposite alteration trend led to negligible contributions (<0.2%) to the routine MTRasym. Meanwhile, CEST effect at 2 ppm and APT presented significantly greater alterations in ADC and CBF lesions compared to ADC/CBF lesion mismatch. Importantly, APT reduction was shown to approximately two-fold of that at 2 ppm in all the three lesion regions, demonstrating the dominant contribution of APT effect to commonly observed pH-sensitive MTRasym change during acute stroke.Discussion and conclusion
APT imaging has been shown capable of dividing perfusion/diffusion mismatch into regions of benign oligemia and metabolic penumbra, refining ischemic tissue delineation. We quantified multi-pool contribution to elucidate the contrast mechanism of MTRasym in acute ischemia. Results confirmed the changes in MT and NOE offset each other, and pH-sensitive amide proton transfer effect dominates the commonly used MTRasym effect.