Vertebral instability in non-neurologic thoracolumbar fractures: the predictive value of methods of measurement.

It is the purpose of this study to evaluate the radiographic results of the conservative treatment of fractures of the thoracolumbar passage in relation to the initial and long-term injury parameters measured by using three different methods: the sagittal index (SI), by Farcy et al., the McRae Index, the Knight et al. Index. The predictive value for each single method was calculated, to determine which of them could be reliable for the purposes of defining primary stability of the injured spine segment. The three methods of measurement were used for initial evaluation and at follow-up (mean 31 months; range 9-45) in 60 non-neurologic thoracolumbar fractures treated conservatively. The fractures were classified based on Magerl et al. Initial instability was evaluated on the basis of the progression of the kyphotic deformity (by at least 5 Cobb degrees as compared to the pre-treatment condition). The Pearson chi 2 test was used for a statistical analysis of the data. There was progression in kyphosis of the fracture in 36 cases (60%). The sagittal index (SI) measured on initial X-rays revealed 27 (76%) potentially stable fractures, while the McRae Index considered 19 to be at risk (55%), the Knight Index, 15 (43%). The higher predictive value (73%) of the sagittal index shows the greater reliability of this method of measurement, as compared to the McRae Index (predictive value = 60%) and the Knight Index (53%) in recognizing the initial phases to be those where the potential instability of the fracture is observed. Even when the SI is applied, the probability of an error in evaluation occurring is 27%. This fact once again stresses the need for more accurate radiological examinations (CT scan, MRI) that clarify the morphological aspects of the lesion injury in an exact manner.