Diffusional kurtosis imaging as a possible prognostic marker of cervical incomplete spinal cord injury outcome: a prospective pilot study.

Abstract

Spinal cord injury (SCI) is associated with substantial chronic morbidity and mortality. Routine imaging techniques such as T1- and T2-weighted magnetic resonance imaging (MRI) are not effective in predicting neurological deficiency grade or outcome. Diffusional kurtosis imaging (DKI) is an MR imaging technique that provides microstructural information about biological tissue. There are no longitudinal prospective studies assessing DKI metrics in acute traumatic SCI. Therefore, the purpose of this study was to establish a DKI protocol for acute SCI and correlate the DKI metrics to the functional neurological outcome of the patients. Eight consecutive SCI patients referred to our institution with cervical SCI were included in the study. An acute diagnostic MRI scan was supplemented with a novel fast, mean kurtosis DKI protocol, which describes the average deviation from Gaussian diffusional along nine different directions. Mean kurtosis values were measured at the injury site and normalized to the mean kurtosis values of a non-injured site. At discharge form specialized rehabilitation, patients were evaluated using the Spinal Cord Independence Measure-III (SCIM-III). The DKI metrics and SCIM-III were analysed using Spearman's rank correlation. This pilot study found a significant correlation between decreasing mean kurtosis values at the injury site of the spinal cord and higher grade of disability measured by the SCIM-III (p = 0.002). This pilot study found that DKI may be a valuable tool as a prognostic marker in the acute phase of SCI.