Relationship between Spinal Cord Volume and Spinal Cord Injury due to Spinal Shortening

Vertebral column resection is associated with a risk of spinal cord injury. In the present study, using a goat model, we aimed to investigate the relationship between changes in spinal cord volume and spinal cord injury due to spinal shortening, and to quantify the spinal cord volume per 1-mm height in order to clarify a safe limit for shortening. Vertebral column resection was performed at T10 in 10 goats. The spinal cord was shortened until the somatosensory-evoked potential was decreased by 50% from the baseline amplitude or delayed by 10% relative to the baseline peak latency. A wake-up test was performed, and the goats were observed for two days postoperatively. Magnetic resonance imaging was used to measure the spinal cord volume, T10 height, disc height, osteotomy segment height, and spinal segment height pre- and postoperatively. Two of the 10 goats were excluded, and hence, only data from eight goats were analyzed. The somatosensory-evoked potential of these eight goats demonstrated meaningful changes. With regard to neurologic function, five and three goats were classified as Tarlov grades 5 and 4 at two days postoperatively. The mean shortening distance was 23.6 ± 1.51 mm, which correlated with the d-value (post-pre) of the spinal cord volume per 1-mm height of the osteotomy segment (r = 0.95, p < 0.001) and with the height of the T10 body (r = 0.79, p = 0.02). The mean d-value (post-pre) of the spinal cord volume per 1-mm height of the osteotomy segment was 142.87 ± 0.59 mm3 (range, 142.19-143.67 mm3). The limit for shortening was approximately 106% of the vertebral height. The mean volumes of the osteotomy and spinal segments did not significantly change after surgery (t = 0.310, p = 0.765 and t = 1.241, p = 0.255, respectively). Thus, our results indicate that the safe limit for shortening can be calculated using the change in spinal cord volume per 1-mm height.