Aims: The purpose of this study was to evaluate the clinical and radiological results of expandable titanium cages for vertebral body replacement in a prospective clinical trial. Methods: Since 04/1999 81 patients with thoracolumbar burst fractures underwent posterior stabilisation followed by vertebral body replacement using expandable titanium cages (VBR, Ulrich, Germany) þlled with cancellous bone graft. Postoperatively at 3, 6, 12 and 24 months clinical and radiological evaluation was performed including ßexion/extension views and quantitative CT-scans to assess stability and fusion. Results: Until 09/2002 40 patients had a one year, 12 a two year follow-up. Pain decreased from 62 to 25 on VAS, ROM increased and preoperative neurologic deþcit improved in 25% of patients. Average postoperative loss of lordosis was 5.5 degrees, subsidence of the cages was 4.5 mm. CT scans showed solid bony fusion in 25%, incomplete fusion in 35% and non-fusion in 40% after one year. One patient suffered of left side paralysis of diaphragm, 9 patients of post-thoracotomy-syndrome. Conclusion: With expandable cages a very exact adaptation to the height of the defect and a gradual press þt of the cage and endplates can be achieved. The clinical outcome after one and two years is similar to operative techniques using tricortical iliac crest bone graft or non-expandable cages. However, fusion could not be achieved in 40% of patients after one year.

Aims: Various studies could show that computer assisted pedicle screw insertion can reduce pedicle perforation rate. We conducted this study to verify if pedicle screw navigation can also avoid neurological complications. Methods: Within 20 months 112 patients were stabilised with 584 pedicle screws in the thoracolumbar spine (Th1-L5). 333 screws were inserted using a CT-based navigation system, 251 srews with conventional technique. Postoperatively, screw positons were assessed by an independent radiologist using CT-scans. Neurological complications and revision surgery were noted. Results: 47 (14.1%) of navigated screws perforated pedicle wall, 13 (28%) to medial side with 2 screws more than 4 mm. One screw (0.3%) had to be changed due to medial perforation of 6 mm in Th4. In the conventional group 60 (29.9%) screws perforated pedicle wall, 13 (22%) to medial side with 3 screws more than 4 mm. One patient had to be reoperated due to radicular deficit caused by a medial perforated screw in L1 (0.4%)Conclusion: CT-based navigation of pedicle screws decreased pedicle perforation rate significantly. However, considerable medial perforations up to 6 mm could not be avoided entirely. The rate of neurologic complicatons and revision surgery was the same in both computer assisted and conventional group.