Somatosensory-evoked potentials as an indicator for the extent of ultrastructural damage of the spinal cord after chronic compressive injuries in a rat model

ObjectiveSomatosensory-evoked potentials (SEPs) were found to correlate well with the disability and postoperative recovery in patients with cervical spondylotic myelopathy. Yet the exact pathophysiology behind it remains to be elucidated. This study aims to characterise the ultrastructural changes of a chronically compressive spinal cord with various SEP responses in a rat model.MethodsA total of 15 rats were used with surgical implantation of a water-absorbing polymer sheet into the cervical spinal canal on the postero–lateral side, which expanded over time to induce chronic compression in the cord. At postoperative 6 months, the functional integrity of the cords was recorded by SEP responses by comparing injured and non-injured sides, and the ultrastructural integrity was assessed by 7-T magnetic resonance (MR) diffusion imaging, contrast-enhanced micro-computed tomography (μCT) and histological evaluations.ResultsSix rats showed unchanged SEP, and the other nine showed decreased amplitude only (n = 5) or delayed latency (n = 4). The circulation insults of the cords were found among all the rats, showing central canal enlargement, intra-tissue bleeding or increased blood vessels in the central grey matter. Ultrastructural damage was noted in the rats with changed SEP responses, which was suggested by lower fractional anisotropy and higher contrast intensity radiologically and echoed by less myelin stain and cavitation changes histologically. In the animals with delayed latency, the cord showed significant loss of motoneurons as well as gross appearance distortion.ConclusionsThe categorised SEP responses by amplitude and latency could be an indicator for the extent of ultrastructural damage of the spinal cord after chronic compressive injuries.SignificanceThe findings built a solid foundation for SEP application in clinical diagnosis and prognostication of spinal cord injuries.

► Spatial filter based on Probabilistic Independent component analysis (PICA) to isolate Somatosensory-evoked potentials (SEP) related components and enhance the signal-to-noise-ratio (SNR) of event related potentials (ERPs). ► Time-frequency filter based on continuous Wavelet filtering to significantly enhance the SNR of ERPs. ► Combining PICA and wavelet filtering offers a space-time-frequency filter to provide a reliable estimation of single-trial ERPs.