BACKGROUND AND PURPOSE: Diffusion kurtosis is a statistical measure for quantifying the deviation of the water diffusion profile from a Gaussian distribution. The current study evaluated the time course of diffusion kurtosis in patients with cerebral infarctions, including perforator, white matter, cortical, and watershed infarctions. MATERIALS AND METHODS: Subjects were 31 patients, representing 52 observations of lesions. The duration between the onset and imaging ranged from 3 hours to 122 days. Lesions were categorized into 4 groups listed above. Diffusion kurtosis images were acquired with b-values of 0, 1000, and 2000 s/mm2 applied in 30 directions; variables including DWI signal, ADC, fractional anisotropy, radial diffusivity, axial diffusivity, radial kurtosis, and axial kurtosis, were obtained. The time courses of the relative values (lesion versus contralateral) for these variables were evaluated, and the pseudonormalization period was calculated. RESULTS: Diffusion kurtosis was highest immediately after the onset of infarction. Trend curves showed that kurtosis decreased with time after onset. Pseudonormalization for radial/axial kurtosis occurred at 13.2/59.9 days for perforator infarctions, 33.1/40.6 days for white matter infarctions, 34.8/35.9 days for cortical infarctions, and 34.1/28.2 days after watershed infarctions. For perforator infarctions, pseudonormalization occurred in the following order: radial kurtosis, ADC, axial kurtosis, and DWI. CONCLUSIONS: Diffusion kurtosis variables in lesions increased early after infarction and decreased with time. Information provided by diffusion kurtosis imaging, including axial and radial kurtosis, seems helpful in conducting a detailed evaluation of the age of infarction, in combination with T2WI, DWI, and ADC.
ASJC Scopus subject areas
- Radiology Nuclear Medicine and imaging
- Clinical Neurology