Magnetic resonance imaging in the prone position and the diagnosis of tethered spinal cord.

OBJECTIVE Tethered cord syndrome (TCS) is a neurosurgical disorder with varied clinical manifestations believed to result from vascular compromise due to stretch forces on the spinal cord. Conventional supine MRI findings may include a low-lying conus medullaris, thickened or fat-infiltrated filum terminale, or lipoma; however, imaging sensitivity and specificity for tethered cord can be low. The purpose of this study was to evaluate the utility of prone MRI in the diagnosis of tethered and retethered spinal cord. METHODS Medical records were reviewed in 41 patients who underwent surgical release of tethered cord and in whom preoperative prone MRI sequences were available. Patients were divided into Group 1 (new TCS diagnosis) and Group 2 (recurrent TCS after previous untethering). Absolute conus ventral motion and motion as a percentage of canal width between supine and prone positions was measured in these 2 groups via sagittal T2-weighted sequences; these groups were compared with 30 consecutive patients (Group 3) who were classified as the normal control group. RESULTS The mean ventral motion was as follows: Group 1 (absolute: 0.5 +/- 0.5 mm [range 0-2.4 mm]; canal percentage: 3.7% +/- 3.9% [range 0%-16.3%]); Group 2 (absolute: 0.4 +/- 0.7 mm [range 0-2.6 mm]; canal percentage: 2.2% +/- 3.7% [range 0%-14.0%]); and Group 3 (absolute: 3.4 +/- 1.3 mm [range 1.4-5.6 mm]; canal percentage: 22.0% +/- 7.2% [range 10.5%-36.1%]). Whereas 38/41 surgically treated patients with TCS had diminished (< 10% canal width) ventral motion on preoperative MRI, 30/30 controls had > 10% canal width motion. Sensitivity and specificity were thereby calculated as 92.7% and 100%, respectively. CONCLUSIONS In the present series, prone imaging is found to be a sensitive and specific tool, and the authors believe it may have a role as supportive evidence in the diagnosis of tethered and retethered spinal cord.