Intervertebral Displacement of the Thoracic Spine with and without Loading: Radiographic and in Vitro Measurements.

Abstract

We are developing an intradural approach to spinal cord stimulation, where the thin electrode array is affixed stably to the underside of the thoracic spinal dura mater without leakage of cerebrospinal fluid. As part of the design and testing process, we sought to evaluate the potential risk of inadvertent contact of the array with the pial surface of the spinal cord during variations in spinal loading. As part of the risk assessment process, a 2-part study was undertaken. First, a retrospective review of the imaging studies of 25 patients was done in the supine, 45- and 90-degree positions to measure the positional shift between the T9 and T10 vertebral bodies as a function of spinal angulation. Second, similar measurements were made on a cadaveric model, with and without a prototype intradural stimulator implanted at the T9-T10 position and with and without 13.8 kg (30 lb) of axial spinal loading at the 90-degree orientation. In all cases, the measured relative displacement of the dura mater towards the spinal cord in both the imaging and the cadaveric arms of the study was less than 1 mm. The implantation method for the thin intradural array of the prototype device will ensure that the anatomic separation between it and the pial surface of the spinal cord will be the same as that of the dura mater. Therefore the risk of inadvertent contact will be no greater than that due to the mass effects of standard epidural stimulator implants.