Could extended laminectomy effectively prevent spinal cord injury due to spinal shortening after 3-column osteotomy?

Objective To explore whether the laminectomy extension can effectively prevent spinal cord injury (SCI) due to spinal shortening after 3-column osteotomy in goat models. Methods A total of twenty healthy goats were included and done with 3-column osteotomy of T13 and L1 under the somatosensory evoked potential (SSEP) monitoring. The samples were divided into two groups. The first group underwent regular laminectomy while the second group underwent an extended laminectomy with an extra 10 mm-lamina cranial to L2. The SSEP measured after 3- column osteotomy was set as the baseline, and the SSEP decreased by 50% from the baseline amplitude and/or delayed by 10% relative to the baseline peak latency was set as positive results, which indicated spinal cord injury. The vertebral column was gradually shortened until the SSEP monitoring just did not show a positive result. The height of the initial osteotomy gap (the distance from the lower endplate of T12 to the upper endplate of L2), the shortened distance (Delta H), the number of spinal cord angulated and the changed angle of the spinal cord (Delta alpha) were measured and recorded in each group. Neurological function was evaluated by the Tarlov scores on day 2 postoperatively. Results All the goats except one of the first group due to changes in the SSEP during the osteotomy were included and analyzed. In the first group, the height of the initial osteotomy segment and the safe shortening distances were 61.6 +/- 2.6 mm and 35.2 +/- 2.6 mm, respectively; the spinal cord of 5 goats was angulated (46.4 +/- 6.6 degrees), the other four goats were kinked and not angulated. In the second group, the height of the initial osteotomy segment and the safe shortening distances were 59.8 +/- 1.5 mm and 43.3 +/- 1.2 mm, respectively, and the spinal cord of ten goats were angulated (97.6 +/- 7.2 degrees). There was no significant difference in the height of the initial osteotomy segment between the two groups by using Independent-Samples T-Test, P = 0.095 (P > 0.05); there were significant difference in the safe shortening distance and the changed angle of the spinal cord between the two groups by using IndependentSamples T-Test (both Delta H and Delta alpha of P < 0.001), the difference between their mean were 8.1 mm and 51.2 degrees. Significant difference was found in the number of spinal cord angulation between the two groups through Fisher's exact test (5/9 vs. 10/10, P = 0.033). Conclusions An additional resection of 10 mm-lamina cranial to L2 showed the satisfactory effect in alleviating SCI after 3-column osteotomy. Timely and appropriate extend laminectomy could be a promising therapeutic strategy for SCI attributable to facilitating spinal cord angulation rather than spinal cord kinking and increasing the safe shortening distance.