IntroductionPrevious studies have found that some patients with wake-up stroke still have ischemic penumbra. ^{[1, 2]} Amide-proton transfer weighted (APTw) imaging can detect the changes in acid-base metabolism in ischemic stroke tissues. ^[3-5] Previous studies mainly focused on patients with cerebral infarction in the hyperacute or acute phase. This study aimed to explore the additional value of the APTw technique to evaluate ischemic penumbra in wake-up stroke patients.
Materials and Methods
Local Ethics Committee approved this Prospective study, and informed consent was obtained from all patients. Ninety-six wake-up stroke patients who underwent routine MRI scans, ASL, and APTw imaging were assessed.
Before the clinical intervention, all patients had undergone conventional and two-dimensional (2D) APTW sequences on a 3.0 T MRI scanner (Ingenia 3.0T; Philips Medical Systems, Best, The Netherlands). A body coil for radiofrequency (RF) transmission and a 16-channel head coil were used for signal reception. The APTw scan parameters were of sensitivity-encoding factor =2, TR=3s, TE=11ms, FOV=240×240 mm², section thickness was 6 mm. Scan time was approximately 3 mins; Conventional MRI sequence include T1-weighted (TR=400ms, TE=20ms), T2-weighted (TR=2800ms, TE=105ms), fluid-attenuated inversion recovery (FLAIR) (TR=8000ms, TE=204ms, TI=2200ms), FOV=240×240 mm², slice thickness=5mm, slice spacing=2 mm, and matrix=512×512.
The APTw raw data were analyzed using interactive data language written in the program (IDL; Research Systems, Inc., Boulder, CO, USA). All cases were divided into the ASL-DWI mismatch group and APTw-DWI mismatch group based on whether the CBF, APTw, and DWI fusion images were matched before treatment. Three regions were delineated on the fusion images of all patients with ischemic penumbra and respectively defined as Infarct core (IC）, mismatch_metabolism , and mismatch_{flow-metabolic}. Five ROIs were randomly delineated for each region to measure APTw values, including the maximum APTw value (APTw_max), minimum APTw value (APTw_min), average APTw value (APTw_ave), and max-min APTw value( APT_max-min).
MRI reexamination after 3 months was used as the standard, and Fisher's exact test was used to compare the sensitivity, specificity, and accuracy of the two mismatch models in detecting IP. Paired t-tests were used to compare APTw parameters between IC, mismatch_metabolic, and mismatch_{flow-metabolic} regions and contralateral ordinary white matter. Kruskal-Wallis test was used to compare the differences in APTw_min, APTw_ave and APTw_max parameters among IC, mismatch_metabolism , mismatch_{flow-metabolic} parts. P<0.05 was considered statistically significant.
Results and discussion
According to the follow-up results, 50 patients had ischemic penumbra, 56 cases were detected in the ASL-DWI mismatch group, and 46 cases were detected in the APTw-DWI mismatch group. The specificity and accuracy of the APTw-DWI mismatch group for IP detection were higher than those of the ASL-DWI mismatch group (100% vs.57%，93.7% vs.90.6%, P＜0.001), but the sensitivity was lower than that of ASL-DWI mismatch group（92%vs.100%）. In the APTw images of 46 patients, APTw_max、APTw_min、APTw_ave in IC region（-0.36（-0.59,-0.17）、-0.89±0.09、-0.59（-1.03，-0.24））and mismatch_metabolic region（0.22（-0.05,0.49）、-0.11±0.10、0.02（-0.25，0.25））were lower than APTw_CNAWM（0.66（0.45，0.82））, and the values in IC region were significantly lower than those in mismatch region（P＜0.001）. APT_max-min in mismatch_metabolic region were significantly higher in mismatch_{flow-metabolic} region ( P < 0.001), However, there was no significant difference in IC region(P＞0.05).
It was found that the APTw-DWI mismatch model had a good consistency with ASL-DWI in IP detection. However, its scope was often smaller than that of the ASL-DWI mismatch model. Moreover, 46 patients also showed a mismatch between perfusion and APTw, and it was found that the APTw value in this region was higher than that in the mismatch_metabolic region and IC region. Compared with DWI-ASL mismatch, APTw can predict and define the ischemic penumbra more accurately. In addition, quantitative studies showed that tissue acidosis in the mismatch_metabolism area was more severe than that in the contralateral normal white matter but less severe than that in the infarct core. The APTw_max-min value of the mismatch_metabolism region was higher than that of the IC region and the mismatch_{flow metabolism} region, indicating that the pH difference was the largest in this area and that the inhomogeneity of APTw signal can be used to discriminate ischemic severity in stroke.
Conclusion
In summary, The APTw-DWI mismatch model is promising for evaluating the ischemic penumbra in patients with wake-up stroke, and its specificity and accuracy are higher than the ASL-DWI mismatch model.

References

[1] Sun PZ, Zhou J, Sun W, et al. Detection of the Ischemic Penumbra Using pH-Weighted MRI[J].Journal of Cerebral Blood Flow & Metabolism. 2006, 27(6): 1129-36. DOI:10.1038/sj.jcbfm.9600424.

[2] Guo C, Bai Q, Zhao Z, et al. Recombinant Tissue-Type Plasminogen Activator Study of Wake-Up Ischemic Strokes Guided by Rapid MRI[J].Cerebrovascular Diseases. 2019, 48(1-2): 85-90. DOI:10.1159/000503379.

[3] Huang D, Li S, Dai Z, et al. Novel gradient echo sequence-based amide proton transfer magnetic resonance imaging in hyperacute cerebral infarction[J].Molecular Medicine Reports. 2015, 11(5): 3279-84. DOI:10.3892/mmr.2015.3165.

[4] Heo HY, Tee YK, Harston G, et al. Amide proton transfer imaging in stroke[J].NMR in Biomedicine. 2022, 36(6). DOI:10.1002/nbm.4734.

[5] Yu L, Chen Y, Chen M, et al. Amide Proton Transfer MRI Signal as a Surrogate Biomarker of Ischemic Stroke Recovery in Patients With Supportive Treatment[J].Frontiers in Neurology. 2019, 10. DOI:10.3389/fneur.2019.00104.