The risk factors of neurological monitoring events in severe thoracic vertebral column resection cases

Abstract

Objective To evaluate the effect of multimodality intraoperative neurophysiological monitoring (IONM) for severe thoracic posterior vertebral column resection (PVCR) surgery, and detail the risk factors of intraoperative monitoring events at various surgery procedures. Methods Monitoring data at various surgery procedures in 82 severe thoracic posterior vertebral column resection cases treated in our center between January 2010 and March 2015 were reviewed. This series of cases including 37 female and 45 male, the initial age averaged 22.6±9.6 years (range 8-66 years). All patients were followed up for minimum 1 year (average 40.7±18.2 months, range 12-74 months). The etiologies of spinal deformity were idiopathic in 36 cases, congenital in 20 patients, tuberculosis in 3 Patients, neurofibromatosis in 9 patients, and neuromuscular in 13 cases. The function of spinal cord and nerve root were assessed by SEP+MEP+ DNEP and tEMG+sEMG combined monitoring during operation, respectively. The monitoring outcomes and the risk factors of cord monitoring events were recorded and analyzed, and then prevention measures were suggested according to our clinical practice and literature review. Results In these cohort cases, combined monitoring of SEP+MEP+DNEP+tEMG+sEMG was successfully achieved in all 82 cases. On average, the major coronal curve was corrected from a preoperative 122.3°±27.6° to a postoperative 55.1°±21.5°, with a correction rate of 55.4%. The sagittal kyphosis was corrected from a preoperative 123.4°±27.8° to a postoperative 53.8°±19.7°, with a correction rate of 56.2%. The average resection vertebra was 1.33±0.6 levels with 11 meshes and 18 cages being planted into 29 patients. The average intra-operative blood loss and surgery time were 2911±1358 ml and 505±105 mins. There were 39 monitoring changes occurred in 27 cases out of 82 (32.9%). Results indicated that neurological monitoring events were more likely to occur in patients with larger scoliosis and kyphosis, longer closure distance of osteotomy, more Halo gravity traction, more screw insertion and higher PVCR segments. The reasons of monitoring changes included screw malposition (6/39), hypotension (2/39), osteotomy maneuver (5/39), cord concussion (4/39), bony compression (7/39), subluxation (4/39), styptic gauze compression (3/39), cord kniking (4/39), and over correction (4/39). After timely detection and intervention during operation, 11 of 82 cases presented new post-operative neural deficit, including 1 incomplete paralysis, 8 transient cord deficit and 2 nerve root injury. Conclusion IONM is effective to predict the spinal cord deficit at different surgery stages. Our report showed PVCR and PVCR site closure had a relatively high neurological risk, and operative manipulation was the main reason of monitoring events. Key words: Thoracic vertebrae; Spinal curvatures; Osteotomy; Evoked potentials, somatosensory; Evoked potentials, motor; Electromyography