The usefulness of diffusion-weighted readout-segmented EPI and fast spin echo with BLADE (PROPELLER) k-space sampling: A comparison with single-shot EPI for diffusion-weighted imaging in ischemic stroke patients

We evaluated the value of diffusion-weighted readout-segmented EPI (DW rs-EPI) and diffusion-weighted fast spin echo images with BLADE (PROPELLER) k-space sampling (DW BLADE) in patients with ischemic stroke through intra-individual comparisons with diffusion-weighted single-shot EPI (DW ss-EPI). Forty-four patients with acute ischemic stroke who had undergone three different types of diffusion weighted imaging (DWI) (b = 1000 s/mm2) were included. Two reviewers assessed the three types of DWI for diagnostic confidence based on a five-point confidence score (CS). The inter- and intra-observer agreements for the CS were assessed. The DWI signal ratio to the contralateral normal area and the apparent diffusion coefficient (ADC) of DWI lesions were compared. The CSs of the three DWI methods were significantly different (P < 0.001). A post hoc analysis showed a significantly higher CS with DW rs-EPI than with DW ss-EPI (P = 0.022). There were significant increases in the CSs for DW rs-EPI compared to DW ss-EPI, for patients who underwent DWI within 24 h of symptom onset, and for patients with DWI lesions < 50 mm2. The inter-observer and intra-observer agreements of the CS were good. DW rs-EPI increased the diagnostic confidence of ischemic stroke compared to DW ss-EPI, particularly for patients with small lesions and for those who underwent DWI within 24 h of symptom onset. DW BLADE showed a similar level of diagnostic confidence to DW ss-EPI.