Fluid-attenuated inversion-recovery (FLAIR) imaging is a technique that produces heavily T2-weighted CSF-nulled images by coupling an inversion pulse followed by long inversion time (TI) to a long echo time (TE) readout. With nulling of the CSF, a tissue abnormality usually becomes the brightest object in the image, thereby improving lesion detection in the brain. The FLAIR technique is also easily adapted to echo-planar imaging (EPI), the most rapid MR imaging technique available. We examined EPI-FLAIR imaging in patients with brain disorders and compared the results with those of turbo-FLAIR and turbo SE T2-weighted imaging. MR imaging was performed on a 1.5 T imager in 29 patients with cerebral infarction and 2 patients with multiple sclerosis. The turbo-SE T2-weighted sequence parameters used were: TR/TE = 4000 ms/99 ms, total scan time = 2 min 12 sec. Turbo-FLAIR sequence parameters were TR/TE/TI = 9000 ms/119 ms/2200 ms, total scan time = 4 min 3 sec. EPI-FLAIR sequence parameters were TE/TI = 60 ms/2200 ms, total scan time = 4.38 sec. EPI-FLAIR images were compared quantitatively and qualitatively with both turbo-FLAIR and turbo-SE T2-weighted images. In the quantiative comparisons of EPI-FLAIR images with turbo-FLAIR and turbo-SE T2-weighted images, lesion-to-white matter contrast and the C/N ratio of EPI-FLAIR images were found to be significantly inferior to both turbo-FLAIR and turbo-SE T2-weighted images (P < 0.001). In the qualitative comparisons, the lesion detection and conspicuty of EPI-FLAIR images were inferior to those of turbo-FLAIR and turbo-SE T2-weighted images. In conclusion, the image quality of EPI-FLAIR images was inferior to that of both turbo-FLAIR and turbo SE T2-weighted images. EPI-FLAIR should not replace turbo-FLAIR.