The spinal cord (SC) has previously been shown to be affected in neurological disorders such as multiple sclerosis (MS) [1], and amyotrophic lateral sclerosis (ALS) [2].

Techniques based on Magnetisation Transfer (MT) imaging provide markers for both brain and spinal cord pathology and the MT ratio (MTR) has previously been shown to be decreased in the SC in MS [4,5], and spinal cord injury (SCI) [6].

However, there are some technical challenges associated with making quantitative MRI measurements in the SC in vivo due to its small cross-sectional area and the potential for SC motion (both physiological and bulk motion) during scans. Differences in physiological motion at the lumbar compared to the cervical level pose further challenges; for example motion may require the use of regional saturation (REST) slabs, which could influence the measurable MTR.

In order to minimise acquisition time and control for distortions near the vertebrae surrounding the SC i.e. bone and tissue/cerebrospinal fluid (CSF) interface, we evaluated the use of single-shot ZOOM-EPI (zonally magnified oblique multi-slice-EPI) [7-9] for lumbar MTR measurement in healthy subjects. ZOOM-EPI is based on an inner volume (IV) imaging technique and makes use of a reduced field-of-view (FOV) and thus shorter echo train length, thereby reducing artefacts caused by susceptibility differences between soft tissue and adjacent vertebrae.

Subjects: Nine healthy volunteers were recruited (4 male, 5 female, mean age 28.5 ± 7.34 years). Five subjects were scanned again with the same protocol between 7 and 14 days following the first scan to assess scan-rescan reproducibility.

MRI acquisition: A 3T Philips Achieva MRI system and the manufacturer’s product 15-channel SENSE spine coil was used, with dual-transmit. Subjects were positioned supine with a wedge foam pad under the knees to reduce curvature of the spine, and to maximise SNR by increasing contact with the spine coil.

A T₂-weighted image of the lumbar spine in the sagittal plane was first obtained and used to facilitate prescription of the slices perpendicular to the spinal cord, positioned at the T11-L1 level.

MTR imaging was performed using a single-shot ZOOM-EPI sequence with the following parameter details: TE/TR=18/7600ms, with and without a train of 40 Sinc-Gaussian shaped MT saturation pulses, each with nominal flip angle=700°, offset frequency=1kHz, duration 20ms (1ms gap), SENSE factor=1.3 in the foot/head direction, half-scan factor 0.612, IV refocusing pulse slice thickness=30mm. Ten 5mm axial slices were acquired, with an in-plane resolution of 1x1mm², reconstructed to 0.5x0.5mm². The acquisition time for both MTon and MToff data (18 signal averages) was approximately 14minutes.

Image Analysis: MTon and MToff images were corrected for motion using slice-wise linear registration implemented in FSL (http://www.fmrib.ox.ac.uk/fsl/), with both MTon and MToff images registered to a mid-point reference image.

The lumbar spinal cord was segmented using the active surface model segmentation method (on co-registered MToff images) in JIM 6.0 (Xinapse systems, www.xinapse.com) [10], with manual editing where necessary, and eroded to remove potential partial volume effects with CSF.

The mean cord MTR value for 9 subjects was 38.8 (±4.05) pu. The mean scan-rescan coefficient of variation (COV (%)) for measuring cord MTR (from 5 subjects) was 4.39%.

Single slice example MToff, MTon and MTR map images are shown in Figure 1.

The COV for MTR measurement in 9 subjects (10.5%) and scan-rescan COV (4.39%) were similar to those measured in a recent pilot study of MTR measurements made using a sequence with a radial acquisition profile performed on three 5mm slices of the cord at the level of the lumbosacral enlargement in healthy volunteers [11]. The radial sequence also allowed tissue-specific (i.e. GM and WM) MTR measurements within the cord but suffered from some disadvantages including loud acoustic noise, high SAR deposition levels and a longer acquisition time than the sequence used here (20 minutes).

The use of ZOOM-EPI is valuable in controlling for extensive physiological motion artefacts associated with the lumbar level of the spinal cord (since it is a single-shot technique) without the need for REST slabs.

The ZOOM-EPI MTR technique provides contiguous-slice, reduced-FOV images in the lumbar SC that do not suffer from aliasing and have reduced magnetic susceptibility artefacts with good inter- and intra-subject reproducibility.

Future work may include investigating the feasibility of obtaining tissue-specific MTR measurements in the lumbar cord, by co-registering high-resolution anatomical scans with MTR data, and assessing the application of the ZOOM-EPI MTR measurement method in neurological diseases affecting the lumbar spinal cord.