Synthetic T2 mapping is correlated with time from stroke onset: a future tool in wake-up stroke management?

Objectives FLAIR-DWI mismatch is an effective method to select eligible wake-up stroke (WUS) patients for intravenous thrombolysis, but shows limitations in the case of subtle FLAIR hyperintensities. T2 mapping is a quantitative method, directly generated from synthetic MRI, which provides T2 relaxation times. We aimed to assess the correlation between T2 values and onset time in acute stroke patients. Methods We prospectively included stroke patients in the 4.5-h window undergoing brain MRI including MAGnetic resonance Image Compilation (MAGiC) from March to October 2017. T2 relaxation times and FLAIR signal intensities were measured in ischemic and contralateral nonischemic regions to calculate FLAIR signal intensity ratio (rSI), difference, and ratio of T2 values. Correlation analysis with time from the onset was achieved using Pearson or Spearman correlation coefficient (rho) test. Results Forty-two patients were included. The strongest correlation with the time from onset was the difference in T2 relaxation times (rho = 0.71; CI95% = [0.48; 0.85]), followed by the ratio (rho = 0.65; CI95% = [0.37; 0.82]) and the absolute T2 relaxation time (rho = 0.4; CI95% = [0.06; 0.66]), whereas the FLAIR rSI showed the weakest correlation (rho = 0.18; CI95% = [-0.16-0.51]). Conclusions The difference and ratio in T2 relaxation times were correlated with the onset time in stroke patients in the 4.5-h window. T2 mapping generated from synthetic MRI may become a relevant tool to select WUS patients with subtle FLAIR hyperintensities. Given that no definitive statement can be made about its usefulness in the 4.5-h windows, further study including patients with an onset time > 4.5 h is required. Key Points The difference and ratio in T2 relaxation times are each individually correlated with the time from stroke onset in the 4.5-h window. FLAIR rSI showed a poor correlation with the time from stroke onset. T2 mapping, directly generated from synthetic MRI, may be a suitable quantitative marker to select safely WUS patients with subtle FLAIR hyperintensities for intravenous thrombolysis.