[Surgical treatment of injuries of the thoracolumbar transition. 2: Operation and roentgenologic findings].

The authors report on a prospective multicenter study with regard to the operative treatment of acute fractures and dislocations of the thoracolumbar spine (T10-L2). The study should analyze the operative methods currently used and determine the results in a large representative collective. This investigation was realized by the working group "spine" of the German Trauma Society. Between September 1994 and December 1996, 682 patients treated in 18 different traumatology centers in Germany and Austria were included. Part 2 describes the details of the operative methods and measured data in standard radiographs and CT scans of the spine. Of the patients, 448 (65.7%) were treated with posterior, 197 (28.9%) with combined posterior-anterior, and 37 (5.4%) with anterior surgery alone. In 72% of the posterior operations, the instrumentation was combined with transpedicular bone grafting. The combined procedures were performed as one-stage operations in 38.1%. A significantly longer average operative time (4:14 h) was noted in combined cases compared to the posterior (P < 0.001) or anterior (P < 0.05) procedures. The average blood loss was comparable in both posterior and anterior groups. During combined surgery the blood loss was significantly higher (P < 0.001; P < 0.05). The longest intraoperative fluoroscopy time (average 4:08 min) was noticed in posterior surgery with a significant difference compared to the anterior group. In almost every case a "Fixateur interne" (eight different types of internal fixators) was used for posterior stabilization. For anterior instrumentation, fixed angle implants (plate or rod systems) were mainly preferred (n = 22) compared to non-fixed angle plate systems (n = 12). A decompression of the spinal canal (indirect by reduction or direct by surgical means) was performed in 70.8% of the neurologically intact patients (Frankel/ASIA E) and in 82.6% of those with neurologic deficit (Frankel/ASIA grade A-D). An intraoperative myelography was added in 22% of all patients. The authors found a significant correlation between the amount of canal compromise in preoperative CT scans and the neurologic deficit in Frankel/ASIA grades. The wedge angle and sagittal index measured on lateral radiographs improved from -17.0 degrees and 0.63 (preoperative) to -6.3 degrees and 0.86 (postoperative). A significantly (P < 0.01) stronger deformity was noted preoperatively in the combined group compared to the posterior one. The segmental kyphosis angle improved by 11.3 degrees (8.8 degrees with inclusion of the two adjacent intervertebral disc spaces). A significantly better operative correction of the kyphotic deformity was found in the combined group. In 101 (14.8%) patients, intra- or postoperative complications were noticed, 41 (6.0%) required reoperation. There was no significant difference between the three treatment groups. Of the 2264 pedicle screws, 139 (6.1%) were found to be misplaced. This number included all screws, which were judged to be not placed in an optimal direction or location. In seven (1.0%) patients the false placement of screws was judged as a complication, four (0.6%) of them required revision. The multicenter study determines the actual incidence of thoracolumbar fractures and dislocations with associated injuries and describes the current standard of operative treatment. The efforts and prospects of different surgical methods could be demonstrated considering certain related risks. The follow-up of the population is still in progress and the late results remain for future publication.