|| Checking for direct PDF access through Ovid
The cases of sixty patients in whom a burst fracture of a thoracic or lumbar vertebral body had been treated with posterior instrumentation and arthrodesis less than two weeks after the injury were retrospectively reviewed. Thirty of the patients had an associated laminar fracture. Eleven of the thirty, all of whom had a lumbar fracture and a preoperative neurological deficit, were noted at operation to have dural laceration. In four of the patients who had dural laceration, neural elements were entrapped between the fragments from the laminar fracture. None of the remaining thirty patients who did not have a laminar fracture had dural laceration (p = 0.0002). Univariate and multivariate statistical analysis revealed no significant association of the dural laceration with the patients' age or sex, or with the radiographic characteristics of the spine. There was a significant association between dural laceration and neurological deficit (p = 0.0001). In our series, the presence of a preoperative neurological deficit in a patient who had a burst fracture and an associated laminar fracture was a sensitive (100 per cent) and specific (74 per cent) predictor of dural laceration. The presence of this fracture pattern and an associated neurological deficit also predicted a risk of dural laceration with entrapped neural elements. This information may influence decisions as to whether an anterior or a posterior surgical approach should be used in such patients.