Background and purpose: A method for identifying patients at increased risk for developing secondary hemorrhagic transformation (HT) after acute ischemic stroke could be of significant value, particularly in patients being considered for thrombolytic therapy. We hypothesized that diffusion-weighted MRI might aid in the identification of such patients.
Methods: We retrospectively analyzed 17 patients with ischemic stroke who received diffusion-weighted MRI within 8 hours of symptom onset and who also received follow-up neuroimaging within 1 week of initial scan. The apparent diffusion coefficient (ADC) for each pixel in the whole ischemic area was calculated, generating a histogram of values. Areas subsequently experiencing HT were then compared with areas not experiencing HT to determine the relationship between ADC and subsequent HT.
Results: A significantly greater percentage of pixels possessed lower ADCs (</=550x10(-)(6) mm(2)/s) in HT lesions compared with non-HT lesions (47% versus 19%; P:<0.001). Moreover, >40% of the pixels possessed values </=550x10(-)(6) mm(2)/s in all lesions experiencing secondary HT, compared with <31% of the pixels in the non-HT-destined lesions.
Conclusions: HT-destined stroke regions possess a significantly great percentage of low ADC values than non-HT-destined regions. Early measurement of ADC values may be a useful tool for assessing secondary HT risk.