Adult Scoliosis Correction: A Comparison Of Techniques

2006

BACKGROUND CONTEXT

Rod derotation, in situ rod bending, and direct vertebral translation to the rod with sublaminar wires have all been used to correct adult scoliosis. A new implant based method (without cables or wires) of direct vertebral translation has also been developed. No studies compare curve correction, complications, and patient outcomes between these techniques.

PURPOSE

To compare clinical and radiographic results in patients with adult scoliosis treated with direct vertebral translation compared to patients treated with rod derotation and in situ bending.

STUDY DESIGN/ SETTING

Prospective treatment group compared to retrospective control group.

PATIENT SAMPLE

The control group (15F, 2M) averaged 48 yrs (20–79 yrs). Twelve were idiopathic and 5 were degenerative curves. The direct translation group (40F, 3M) averaged 60 yrs (19–85 yrs). 21 were idiopathic (1 thoracic, 3 thoracolumbar, 8 lumbar, 9 double major), 15 were degenerative, 7 were revisions.

OUTCOME MEASURES: Oswestry and Visual Analog pain Scores (VAS), pain medication use, and work status were followed along with radiographs.

METHODS

Sixty consecutive patients with adult scoliosis treated with posterior correction by one surgeon. The first 17 (control group) were corrected by rod derotation and in situ rod bending and followed 5 yrs (range 3-11 yrs). The next 43 patients (followed 3.5 yrs, range 24-56 mo) were corrected by direct vertebral translation by pulling the spine to a contoured rod via the pivoting reduction posts on the screws. This technique produced correction in both the coronal and sagittal planes. Anterior surgery was required in 15/17 control and 39/43 direct translation patients. Two patients from each group required osteotomies to mobilize ankylosed segments. Clinical and radiographic results were analyzed by curve type.

RESULTS

The direct translation group curves of 49° (17–83°) corrected 72% to 14° (4-40°) was better (P<0.01) than control group curves of 55° (25–84°) corrected 48% to 29° (10-59°). Idiopathic scoliosis of 58° (43–83°) in the translation group corrected 69% to 18° (7–40°) compared to 49% correction in the control group. Correction of translation group vs control group by curve type was: degenerative 67% vs 49%, thoracic 70% vs 51%, thoracolumbar 81% vs 44%, lumbar 74% vs 67%, double major 62% vs 34%. Control group complications included 3 nonunions (17%), 2 cases of implant pullout (11%), 1 broken rod, 1 infection (5%). The translation group had 4 nonunions (9%), 2 infections (5%), no cases of screw pullout during reduction. 11/14 returned to work. Oswestry and VAS scores were similar for both groups, showing significant improvements in pain and function at 1 and 2 years.

CONCLUSION

This study shows statistically improved correction of adult scoliosis by direct vertebral translation using a screw with pivoting reduction posts compared to other techniques. The most dramatic improvement was seen in patients with thoracolumbar and lumbar scoliosis. The technique appears to be very promising in patients with adult scoliosis.