Sensitivity and specificity of in vivo diffusion-weighted MRI in acute spinal cord injury

Abstract

The aim of this study was to test the sensitivity and specificity of diffusion-weighted MRI for the detection of acute spinal cord injury. Forty female New Zealand white rabbits were randomly divided into four groups: the mild, moderate and severe injury groups, and the control (sham operation) group. Contusion of the spinal cord was induced using a weight-drop impactor. All animals were imaged using T1-weighted, T2-weighted, and diffusion-weighted imaging (DWI) sequences at 30 minutes, 6 hours, and 24 hours after injury. One animal from each group was killed at each time point for histologic examination of the spinal cord. DWI had a sensitivity of 100% at all time points, whereas T2-weighted MRI had a sensitivity of 43.33% at 30 minutes after injury, 81.48% at 6 hours after injury, and 95.83% at 24 hours after injury. Conversely, the specificity of DWI was lower than that of T2-weighted MRI at all time points. One animal in the control group had a non-specific high signal on DWI. Significant systematic differences were seen between DWI and T2-weighted MRI at both 30 minutes and 6 hours after injury. The apparent diffusion coefficient values of the lesion were lower than those of adjacent unaffected regions in the mild and moderate injury groups, but higher than adjacent unaffected regions in the severe injury group. The histological findings were reliably correlated with the magnetic resonance findings. We found that DWI has a higher sensitivity, but a lower specificity, than conventional MRI for the detection of early pathological changes after contusive injury.