Diffusion-weighted MR imaging within 24 h post-injury after traumatic spinal cord injury: a qualitative meta-analysis between T2-weighted imaging and diffusion-weighted MR imaging in 18 patients

Abstract

Only few studies have been published about diffusion-weighted imaging (DWI) within 24 h of traumatic spinal cord injury (tSCI). The purpose of this study was to compare the imaging findings from conventional magnetic resonance imaging (MRI) and DWI in seven tSCI patients with findings in the existing literature. Seven patients with tSCI at neurologic levels C2-T10 were examined with conventional MRI and DWI within 24 h post-injury. DWI was obtained with a b-factor of 1000 s mm(-2). American Spinal Injury Association (ASIA) scores and Spinal Cord Independence Measurement (SCIM) II item 12 after 12 months were collected. In addition, MEDLINE was searched from 1995 to 2010 to identify clinical tSCI studies reporting on MRI, DWI and apparent diffusion coefficient maps within 24 h post-injury to perform a meta-analysis. Images obtained with a b-factor of 1000 s mm(-2) were compared with lower b-factors. Differences were calculated using χ (2) tests. No associations were identified between the images of the seven tSCI patients and ASIA or SCIM II scores. Eighteen SCI patients (11 from the retrieved publications) were included in the meta-analysis. The detection rates of hyperintense signals on T2-weighted and DW imaging did not show significant differences at 94 and 72%, respectively. In addition, there were no significant differences in detection rates or diffusion abnormalities between subjects in whom DW images were obtained with a maximum b-factor of 1000 or <1000 s mm(-2). Our analysis suggests that T2-weighted and DW imaging have comparable detection rates for spinal cord damage in tSCI patients within 24 h post-injury.