Cerebral spinal fluid contamination of the measurement of the apparent diffusion coefficient of water in acute stroke

Abstract

The measurement of the apparent diffusion coefficient (ADC) of water in brains of stroke patients is used in models developed to help distinguish reversible from irreversible ischemic injury. The ADC by conventional methods may be overestimated by the presence of cerebral spinal fluid (CSF) in sulci and perivascular spaces. In this study the hypothesis that DWI with CSF suppression (FLAIR-DWI) would result in different ADC values than those obtained with the conventional DWI technique was investigated. Thirty-one patients with stroke onset of less than 6 hr and an acute lesion on conventional DWI were studied. Both conventional isotropic DWI and FLAIR-DWI were performed using a single-shot echo-planar technique. In all 31 patients, CSF-suppressed ADC was lower than conventional ADC. The mean (SD) of the 31 patients' lesion ADC was 0.64 (0.08) x 10(-3) mm(2) s(-1) with FLAIR-DWI and 0.72 (0.09) x 10(-3) mm(2) s(-1) with conventional DWI (P < 0.001). The overestimation of ADC in conventional DWI corresponded to the percentage of the voxel that contained CSF. Suppression of CSF leads to lesion ADC values that are more homogeneous and more than 15% lower than those obtained with conventional DWI techniques. This suggests that FLAIR-DWI ADC measurements are more accurate than conventional ADC maps.