[The effectiveness of short-segment posterior instrumentation of thoracolumbar burst fractures]

We evaluated the results of short-segment posterior instrumentation of thoracolumbar burst fractures and investigated correlations between radiographic and functional results as well as factors that affected correction losses.

We reviewed 48 patients (30 males, 18 females; mean age 40+/-14 years; range 18 to 67 years) who underwent short-segment posterior instrumentation with pedicle screws and fusion. The most common involvement was at L1 in 18 patients (37.5%), followed by T12 in 11 patients (22.9%). According to the Frankel grading system, six patients had complete, 14 patients had incomplete neurologic deficits. The Cobb angles were measured, and canal remodeling was assessed by computed tomography. Modified functional results were derived using the Denis pain and work scales. The mean follow-up was 21.7+/-9.1 months (range 12 to 48 months).

The mean correction in the Cobb angle was 18.2+/-8.6 degrees (p<0.01), the mean correction loss was 7.4+/-5.7 degrees (p<0.01), and the mean canal remodeling was 51.3+/-9.3% (p<0.001). There was a significant correlation between Cobb angle correction and correction loss (r=0.38, p<0.01). An intraoperative correction of greater than 15 degrees was significantly associated with a greater correction loss (p<0.05). Patients with a correction loss of more than 10 degrees had a significantly poorer Denis pain score and modified functional result (p<0.05). Modified functional results were excellent in 16 patients (33.3%), good in 23 patients (47.9%), fair in seven patients (14.6%), and poor in two patients (4.2%). At final follow-ups, the Cobb angle was not correlated with functional results (p>0.05). All the patients having incomplete neurologic deficits improved by at least 1 Frankel grade.

An intraoperative correction exceeding 15 degrees is significantly associated with a greater correction loss, which in turn has a significantly adverse effect on functional results.