Correction of Adolescent Idiopathic Scoliosis Using a Convex Pedicle Screw Technique: A Novel Technique for Deformity Correction.

Background: We describe our convex segmental pedicle screw technique for the treatment of adolescent idiopathic scoliosis. We developed this technique to achieve optimum 3-dimensional deformity correction while reducing the surgical risks of an inherently dangerous procedure.

Description: The surgery involves a wide posterior subperiosteal exposure across the deformity levels to the tips of the transverse processes. Posterior releases are performed through facetectomies. Pedicle screws are placed using a freehand technique based on anatomical landmarks. Adequate screw positioning is assessed with an image intensifier before rod engagement. Segmental pedicle screws are placed across the convexity of each curve included in the fusion. Proximal and distal fixation of the rods on the contralateral side is performed across 2 pedicle screw anchors. We use titanium rods bilaterally. Curve correction is done using the convex pedicle screws by applying segmental vertebral translation and derotation starting with the main thoracic curve followed by the lumbar curve. Segmental compression or distraction is performed at the proximal and distal ends of the construct to level the end vertebrae included in the fusion. Maximum correction of the main thoracic scoliosis is done, whereas the lumbar scoliosis (which is usually more flexible) is corrected to the point that results in a globally balanced spine in the coronal plane. The rod attached on the convex side of the main thoracic scoliosis is overbent to restore thoracic kyphosis, and the aim is always to achieve regional and global sagittal balance. An interfacetal, intertransverse, and interlaminar fusion is performed with use of locally harvested bone supplemented by allograft bone.

Alternatives: With previous techniques, the use of bilateral segmental pedicle screw fixation has been advocated as a requirement to achieve adequate deformity correction in patients with adolescent idiopathic scoliosis.

Rationale: This technique is associated with low risks of neurological and vascular complications because the screws are placed at the convex pedicles, away from the spinal cord/cauda equina and the aorta. The use of far fewer pedicle screws compared with previous techniques reduces surgical time and blood loss, which is related to lower postoperative morbidity. It may also decrease the risk of deep wound infection, which is associated with the number of implants used. Low implant density (1.2, with a density of 2 representing placement of pedicle screws bilaterally at every instrumented segment) with our technique can achieve satisfactory scoliosis correction, improved thoracic kyphosis, and normal global sagittal balance. Our use of this technique has resulted in excellent patient satisfaction and functional outcomes with no neurological complications or intraoperative neuromonitoring events, deep wound infections, detected nonunions, or need for revision surgery.