Introduction: The anatomy and biomechanics of the thoracic spine is different from the cervical and lumbar spine particularly due to the ribs and sternum which contribute to stability and controlling motion. The role of the sternum and costosternal articulation in the biomechanics of thoracic fracture or deformity correction has not been well studied. The effects of releasing each of these structures, whether alone or in combination, is potentially relevant in the surgical correction of thoracic deformities such as severe kyphosis. The purpose of this study was to investigate the relative effects of releasing the intervertebral disc, the costosternal joint, the sternum, and the facet joints on sagittal thoracic motion and the consequences of altering the sequence of the releases.

Methods: Eighteen human torsos were tested in three experiments (A, B, and C) to determine the effect on sagittal motion due to three different sequences of three surgical releases. In Experiment A the release sequence was back to front: Total facetectomy, then radical discectomy, then sternal osteotomy plus costosternal release. In experiment B the release sequence was front to back: Sternal osteotomy plus costosternal release, then radical discectomy, then total facetectomy. In Experiment C, it was disc first: Radical discectomy, then sternal osteotomy plus costosternal release, then total facetectomy. The different sequences allowed separate analysis of each component and the synergistic patterns. In each of the three experiments, the torso was flexed then extended each time by an applied force (25 N) before and after each release. The extent of both angular flex-ion and angular extension were compared to the intact condition, and after each release.

Results: Radical discectomy provided the greatest increase (P< 0.05) in range of motion (ROM) as compared to the other two single releases, no matter what the sequence. For paired release combination, the radical discectomy and sternal osteotomy plus costosternal release (as in Experiments B and C) provided a significant (P< 0.05) increase in sagittal ROM compared to the combination of radical discectomy and total facetectomy (Experiment A). In Experiment A, if sternal osteotomy and costosternal release (the final release) had not been carried out, then 42% of the sagittal motion would have been lost compared to the 27% related to the total facetectomy (Experiment B). All of the releases allowed more extension than flexion; the only exception was facetectomy when carried out first as in Experiment A.

Conclusions: To increase sagittal thoracic range of motion radical discectomy provided the greatest increase in both extension and total ROM as compared to total facetectomy or sternal osteotomy plus costosternal release, no matter what the sequence. For two releases, the combination of radical discectomy and sternal osteotomy plus costosternal release provided the greatest increase in both extension and total ROM. Total facetectomy was the least useful release. These data have relevance for surgical strategies to correct severe thoracic sagittal plane deformity. The sequence of combined release has important clinical implications.