Sagittal Correction and Reciprocal Changes After Thoracic Pedicle Subtraction Osteotomy.

STUDY DESIGN: A retrospective clinical review. OBJECTIVE: To evaluate the sagittal correction efficacy of a thoracic pedicle subtraction osteotomy (PSO), to determine and predict changes at both the cervical and lumbar mobile unfused segments after whether an upper or lower level thoracic PSO. SUMMARY OF BACKGROUND DATA: Thoracic PSO is a technically challenging but increasingly valid procedure in the treatment of fixed thoracic deformities.Anatomical characteristics differentiate upper and lower segment thoracic PSOs and define its corrective ability. There is yet paucity in the literature concerning the causality between the osteotomy level and the reciprocal changes observed postoperatively at the adjacent unfused curves. METHODS: Data collection from a single institutional database was carried out retrospectively by reviewing medical records and imaging of fixed thoracic kyphotic deformity patients submitted to a single level thoracic PSO. RESULTS: Seventeen patients (11M:6F), with a mean age of 44 years (range, 17-76). Fifty-nine percent had a previous spine surgery. The mean follow-up was of 55 months (range, 27-122). In six patients PSO was performed in the upper thoracic segment (T1-T6) and in 11 patients between T7 and T12 (lower thoracic segment), allowing a mean local kyphosis correction of 31° and 41° respectively. The mean cervical and lumbar lordosis globally improved. The upper thoracic PSO group had a significant improvement of the cervical lordosis with a mean C7 slope correction of 28° (P = 0.004), whereas in the lower thoracic PSO group a more pronounced correction of the maximal lumbar lordosis was observed (22°, P = 0.033). CONCLUSION: After a fixed sagittal thoracic deformity, the predominant lordotic compensation seems to occur at the nearest mobile curve from the deformity apex. The PSO-induced restoration of the thoracic kyphosis relieved the compensatory cervical and/or lumbar excessive lordosis, which were present preoperatively and are now no longer needed to achieve balance and maintain horizontal gaze. LEVEL OF EVIDENCE: 4.