CT for thoracic and lumbar spine fractures: Can CT findings accurately predict posterior ligament complex injury?

Abstract

This study aims to determine whether secondary CT findings can predict posterior ligament complex (PLC) injury in patients with acute thoracic (T) or lumbar (L) spine fractures. This is a retrospective study of 105 patients with acute thoracic and lumbar spine fractures on CT, with MRI as the reference standard for PLC injury. Three readers graded CT for facet joint alignment (FJA), widening (FJW), pedicle or lamina fracture (PLF), spinous fracture (SPF), interspinous widening (ISW), vertebral translation (VBT), and posterior endplate fracture (PEF). Univariate and multivariate logistic regression analyses were performed separately for each reader to test for associations between CT and PLC injury, and diagnostic performance of CT was calculated. Fifty-three of 105 patients had PLC injury by MRI. Statistically significant predictors of PLC injury were VBT, PLF, ISW, and SPF. Using these four CT findings, odds of PLC injury ranged from 3.8 to 5.6 for one positive finding, but increased to 13.6-25.1 for two or more. At least one positive CT finding was found to yield average sensitivity of 82% and specificity 59%, while two or more yielded sensitivity 46% and specificity 88%. While no individual CT finding is sufficiently accurate to diagnose or exclude PLC injury, greater the number of positive CT findings (VBT, PLF, ISW, and SPF), the higher the odds of PLC injury. The presence of a single abnormal CT finding may warrant confirmatory MRI for PLC injury, while two or more CT findings may have adequate specificity to avoid need for MRI prior to surgical intervention. These slides can be retrieved under Electronic Supplementary Material.