Michael Y Wang, Gerd Bordon

OBJECTIVE Pedicle subtraction osteotomy (PSO) is a powerful but high-risk surgical technique for destabilizing the spine for deformity correction in both the sagittal and coronal planes. Numerous reports have demonstrated the benefits of this technique for realigning the spine in a physiological posture; however, the open surgical technique is associated with a high complication rate. In this report the authors review data obtained in a series of patients who underwent PSO through a less invasive approach. METHODS Sixteen patients with severe coronal- and/or sagittal-plane deformities were treated in this series. Conservative measures had failed in all cases and patients had undergone a single-level PSO or extended PSO at L-2 or L-3. Fixation was accomplished using percutaneous instrumentation and interbody or facet joint fusions were used at the remaining levels. None of the procedures were aborted or converted to a traditional open procedure. Standard clinical and radiographic measures were used to assess patient outcomes. RESULTS Mean age was 68.8 years and mean follow-up duration was 17.7 months. An average of 7.6 levels were fused, and 50% of the patients had bilateral iliac screw fixation, with all constructs crossing both the thoracolumbar and lumbosacral junctions. Operative time averaged 356 ± 50 minutes and there was a mean blood loss of 843 ± 339 ml. The leg visual analog scale score improved from a mean of 5.7 ± 2.7 to one of 1.3 ± 1.6, and the back visual analog scale score improved from a mean of 8.6 ± 1.3 to one of 2.4 ± 2.1. The Oswestry Disability Index score improved from a mean of 50.1 ± 14.4 to 16.4 ± 12.7, representing a mean reduction of 36.0 ± 16.9 points. The SF-36 physical component summary score changed from a mean of 43.4 ± 2.6 to one of 47.0 ± 4.3, and the SF-36 mental component summary score changed from a mean of 46.7 ± 3.6 to 46.30 ± 3.0. Coronal alignment improved from a mean of 27.9 ± 43.6 mm to 16.0 ± 17.2 mm. The lumbar Cobb angle improved from a mean of 41.2° ± 18.4° to 15.4° ± 9.6°, and lumbar lordosis improved from 23.1° ± 15.9° to 48.6° ± 11.7°. Pelvic tilt improved from a mean of 33.7° ± 8.6° to 24.4° ± 6.5°, and the sagittal vertical axis improved from 102.4 ± 73.4 mm to 42.2 ± 39.9 mm. The final lumbar lordosis-pelvic incidence difference averaged 8.4° ± 12.1°. There were 4 patients who failed to achieve less than or equal to a 10° mismatch on this parameter. Ten of the 16 patients underwent delayed postoperative CT, and 8 of these had developed a solid arthrodesis at all levels treated. A total of 6 complications occurred in this series. There were no cases of symptomatic proximal junction kyphosis. CONCLUSIONS Advancements in minimally invasive technique have resulted in the ability to manage increasingly complex deformities with hybrid approaches. In this limited series, the authors describe the results of utilizing a tissue-sparing mini-open PSO to correct severe spinal deformities. This method was technically feasible in all cases with acceptable radiographic outcomes similar to open surgery. However, high complication rates associated with these deformity corrections remain problematic.