Adolescent Idiopathic Scoliosis Treated with a Posterior All-Pedicle Screw Construct

Lawrence G. Lenke, MD

The Jerome J. Gilden Professor of Orthopedic Surgery

Co-Chief Pediatric & Adult Spinal, Scoliosis & Reconstructive Surgery

History:
The patient is an 18 year-old male who was initially noted to have right thoracic scoliosis at the age of 17 and was followed both clinically and radiographically. At the time of presentation, his curve measured 32 degrees and he was thought to be relatively skeletally mature with a Risser 3-4. At routine one-year follow-up, his parents mentioned that his right shoulder seemed to be getting higher and his ribs stuck out more on the right side. His radiographs showed that his curve had increased in magnitude to 53 degrees and he was clinically more imbalanced with his trunk shifted to the right.

The options for treatment included routine follow-up at regular intervals or surgery. The patient was not thought to be a good candidate for brace treatment as he was judged to be skeletally mature based on radiographic analysis of his growth plates. Because he was fairly imbalanced and with a definite risk of progression throughout adulthood, the patient and his parents opted for surgical treatment.

AP shows a 53-degree R thoracic curve.	Lateral shows a neutral thoracic kyphosis. Note the endplate changes consistent with lumbar Scheurmann's disease.
Standing posture shows an elevated R shoulder and scapular asymmetry.	Forward bend demonstrates the large right rib hump. Scoliometer measurement was 16°.

Surgery:
Based on both the patient's clinical appearance and radiographs, a posterior spinal fusion and instrumentation was thought to be the best option for this case. While both an anterior or a posterior approach are feasible, the posterior approach was thought to be a better choice because of the necessity for instrumentation up to the high thoracic spine as well as the patient's pulmonary function testing being only 60% of predicted.

The patient underwent a posterior spinal instrumentation and fusion from the upper thoracic spine to the upper lumbar spine (T3-L1) with iliac crest bone graft. The levels down to L1 were included in the construct because of the lumbar Scheurmann's disease. Fixation into the spine was achieved with pedicle screws at each level of the left-sided correcting rod. With appropriate contouring, the residual curve after surgery was 8o. Postoperatively the patient was in the hospital for 5 days. He was able to return to school after four weeks.

Standing posture shows symmetry of scapulae.The left shoulder is a bit elevated at 6 weeks; the shoulders were well balanced at the 3-month visit.	Forward bend shows reduction of rib hump.

Discussion:
Traditionally posterior scoliosis instrumentation consists of a hook-rod construct that relies on good purchase of the hooks into the posterior elements of the levels to be fused. While hooks provide a great deal of flexibility in placement, the corrective maneuver can cause hook loosening. If overly aggressive forces are applied to a hook construct, there is the definite possibility of posterior element fracture and hook dislodgement. Therefore, there is a limit to the amount of correction that can be achieved with an all-hook construct based on the maximum force that can be applied to the posterior elements.

A more recent technique that has been gaining popularity is the use of thoracic pedicle screws in scoliosis surgery. Pedicle screws allow purchase into all three columns of the spine, consequently allowing a much greater force to be imparted during the corrective maneuver. As well, pedicle screws remain tightly affixed to the vertebrae during correction without slipping, unlike hooks. In our center, a review of our all-pedicle screw cases showed an average correction of 77% as compared to 49% with an all-hook construct.

Pedicle screws may offer other advantages that will only be recognized with long-term follow-up. Long-term follow-up of hook constructs has demonstrated transition problems below and above the end fusion levels, probably related to the disruption of the posterior soft tissue that acts as a tension band to keep the spine extended. The exposure for pedicle screws causes much less soft tissue disruption at the ends of a surgical construct as opposed to hooks and therefore would be expected to cause fewer problems with transition syndrome. As well there is some thought that secure fixation of all three columns of the spine may reduce the risk of "crankshaft" problems in skeletally immature patients who might other wise need both an anterior and posterior procedure to arrest growth.

There are definite downsides to pedicle screws that must be recognized. Placing pedicle screws in the thoracic spine is a technically demanding operation that requires intimate knowledge of spinal anatomy and extremely thorough exposure of the entire posterior spine. Particularly on the concave aspect of the deformity, the pedicles are quite small and may be difficult to access. Rigorous preoperative planning based on multiple radiographs will help the surgeon identify which pedicles may be too small to place screws. We have found that the supine and "push-prone" x-rays are very helpful in removing some of the rotation of the spine to allow a better view of the pedicles. Fluoroscopy in the OR can also help to identify the pedicles both prior to and during insertion of the screws. A full complement of screws in various sizes is necessary for each case, as there is a great amount of variability in the pedicle sizes at the different levels.

In summary, scoliosis constructs with all-pedicle screws are an important advance in the surgical treatment of scoliosis. They allow more correction to be obtained than with all-hook constructs and have a very low risk of having dislodgement from the spine at either end of the fusion. They may also allow more procedures to be done with just a posterior surgery that would have previously required an anterior procedure as well. This is certainly an evolving area that requires much more work and research.