Posterior cervicothoracic instrumentation in spine tumors.

STUDY DESIGN: We retrospectively review 32 patients who underwent posterior fixation for cervicothoracic junctional tumors. All patients possessed unstable or potential after surgery unstable spines as a result of either their tumors or the surgery performed. We examined cervicothoracic spine stability, maintenance of alignment, and associated complications.

OBJECTIVES: To review our experience with 3 different posterior osteosynthesis systems applied to the cervico-thoracic junction for spinal tumors. Our review includes surgical outcomes and complications. The evolution through 3 different systems between 1994 and 1997 reflects our attempts to improve accuracy in light of variable facet and pedicle interspaces. Our goal is not to compare the efficacy of the systems but to assess the efficiency of cervicothoracic facet and transpedicular screw and plate or rod fixation. However, we will comment on why the evolution occurred. The 3 different systems share a similar characteristic. Each system employs posterior cervical facet screw fixation and thoracic trans-pedicular screw fixation.

SUMMARY OF BACKGROUND DATA: Spinal disorders involving the cervicothoracic junction and specific instrumentation to this region have been sparsely described in the literature.

METHODS: Between June 1994 and June 2000, 32 patients underwent surgery for spinal tumors involving the cervicothoracic junction at our institution. There were 27 males and 5 females. The ages ranged from 17 to 72 years with a mean age of 52 years. A total of 32 cervicothoracic instrumentations were performed. We used the R. Roy-Camille thoracolumbar plate in 20 patients, the cervico-thoracic plate in 8, and the Agora rod system in 4. In all, 96 lateral mass screws were implanted from C4 to C6, 54 into C7, and 180 pedicle screws from T1 to T8. Nineteen patients had lung cancer with vertebral body invasion (Pancoast tumors), 11 had metastasis to the cervicothoracic junction, 1 had a chondrosarcoma, and 1 had myeloma. In a first group consisting of 19 patients, a combination of anterior and extended posterior surgical approaches allowed complete en bloc resection of the tumors, including all invaded vertebrae. Four total vertebrectomies and 15 partial vertebrectomies were performed. A second group of 13 patients had only posterior palliative stabilizing procedures with laminectomy and cervicothoracic fixation.

RESULTS: The follow-up period varied from 3 to 54 months, average 15 months. The average duration of survival for patients who underwent partial or total vertebrectomy was 16 months (range 3-54 months). The average duration of survival for patients who underwent palliative decompression and stabilization was 11 months (range 5-19 months). No changes in the sagittal alignment occurred during the immediate postoperative period for 30 patients. However, 2 mechanical failures occurred. Two patients experienced a clinically significant early increase in thoracic kyphosis and required revision of the posterior instrumentation. A 21-month minimum follow-up was available for 6 patients, in whom all implants were stable. We noted no screw, plate, or rod breakage in this series. No neurologic complications, including root impingement or spinal cord injury, or vertebral artery injury occurred related to screw insertion into either the thoracic pedicles (180 screws) or the cervical lateral masses (96 screws in C4-C5-C6 and 54 screws in C7).

CONCLUSIONS: Posterior plate or rod and screw fixation is a good method of treatment for cervicothoracic instability in spine tumors. Facet screw fixation in the cervical spine with Roy-Camille drilling technique and transpedicular screw fixation in the thoracic spine provides an efficacious means by which to stabilize the cervicothoracic junction. This stabilization technique was effective even in cases of high postoperative instability, such as with partial or total vertebrectomy. This screw-type stabilization is clinically effective and well documented. The evolution through 3 different systems reflects our attempts to improve accuracy in light of variable facet and pedicle interspaces. Importantly, posterior instrumentation will not interfere with subsequent laminectomy or more extreme surgical procedures.