Burst fractures of the thoracolumbar spine: changes of the spinal canal during operative treatment and follow-up.

Although multiple studies have concluded operative decompression of a traumatically narrowed spinal canal is not indicated because of spontaneous remodeling, instrumental decompression is frequently used as part of the operative treatment of spinal fractures. To investigate the process of remodeling, we studied the diameter of the spinal canal in 95 patients with burst fractures at the thoracolumbar junction (T9-L2). To measure and compare the spinal canal's diameter we used either computed tomography (CT) scans or radiographs, made preoperatively, postoperatively, after 9 months and after 24 months. In lateral plain radiographs we found that the initial percentage of cases with bony canal narrowing preoperatively of 76.5 was reduced to 18.4% postoperatively, to 8.2% at 9 months, and to 2.4% at 24 months. In CT scans in a selection of patients, the mean residual diameter of the spinal canal was 53% preoperatively and 78% at 24 months. The posterior segmental height increases during operation and decreases in the respective periods after operation. So ligamentotaxis can only play a role in the perioperative period. We conclude that a significant spontaneous remodeling of the spinal canal follows the initial surgical reduction. Two years after operation, bony narrowing of the spinal canal is only recognizable in 2.4% of the patients on plain lateral radiographs. The remodeling of the spinal canal can be seen on plain radiographs, although not as accurately as on CT scans.

Classificational problems in ligamentary distraction type vertebral fractures: 30% of all B-type fractures are initially unrecognised.

The clinical records, operation records, X-rays and CT-scans of 160 operatively treated patients with A-type and B-type spinal fractures were evaluated in a retrospective study. The preoperative diagnosis was compared with the postoperative diagnosis. Analysis of characteristics of patients with A-type fractures (without the unrecognised B-type fractures), initially unrecognised B-type (uB) fractures, and B-type fractures (without the unrecognised B-type fractures) was performed. We analysed the age of the patients, the respective fracture levels, neurologic deficit, anterior wedge angles (AWA), anterior corporal height (ACH), posterior corporal height (PCH), and the percentage of frontal corporal collapse (FCC). The t-test was used for statistical analysis. The mean age of patients in each group did not show a significant difference. The group of unrecognised B-fractures had a more caudal fracture level than the recognised B-type fractures. The fracture levels of the A-group and the uB-group patients showed no difference using the t-test. The percentage of patients with spinal fractures with neurologic deficit is 16% in the A-type fracture group, 12% in the uB-fracture group and 50% in the B-type group. The preoperative classification of patients in the A-group and in the uB-group showed that patients in the uB-group have more than proportional relatively simple preoperative A-fractures. The AWA and ACH did not show significant differences between the groups. The mean PCH of the uB-group was higher than the PCH of the A-group. No differences were measured between the uB-group and the B-group. The mean percentages of frontal corporal collapse (FCC) did not show a significant difference. Thirty percent of B-type fractures are misdiagnosed when plain X-rays and CT scans with 2D reconstructions are used as the only preoperative diagnostic tools. A large PCH with a normal interspinous distance should raise the suspicion of a B-type lesion. A large AWA does not point to a ligamentary B-type fracture.

Thoracolumbar spinal fractures: segmental range of motion after dorsal spondylodesis in 82 patients: a prospective study.

In order to study the effect of dorsal spondylodesis on intervertebral movement in patients treated for thoracolumbar fractures, we measured the sagittal range of motion (ROM) in the segments above and below the fractured vertebral body 2 years after operation. Between 1991 and 1996, 82 consecutive patients with a fracture of the thoracolumbar spine (T12, L1, L2 and L3) were treated operatively with open reduction and stabilisation using an internal fixator, combined with transpedicular cancellous bone graft and dorsal spondylodesis. Eighteen T12, 42 L1, 17 L2 and 5 L3 fractures were included. The range of motion of two segments above and two segments below fracture level was measured. This was done on plain flexion and extension radiographs. The data were compared to normal values and to the zero distribution with the Kolmogorov-Smimov test. At all fracture levels the ROM of the segment adjacent to the disturbed endplate of the fractured body was zero (K-S test). All other evaluated segments showed significant loss of ROM (P<0.05) compared to normal values, except segment L1-L2 in L3 fractures (P=0.058). Dorsal spondylodesis at the level of the disturbed endplate in thoracolumbar spinal fractures leads to immobility in this segment, measured on flexion-extension radiographs 2 years after primary operative treatment. More than 50% loss of motion in the two adjacent levels is equivalent to complete loss of ROM in a second segment.