Background: Loss of coronal correction has been reported in the presence of a solid posterior fusion mass in patients who are skeletally immature. Significant increases in postoperative coronal curvature are typically attributed to the presence of pseudarthrosis, loss of instrumentation fixation, or adding-on. We evaluated the clinical and radiographic factors associated with a loss of correction that was not attributed to these etiologies as well as the impact of pedicle screw fixation on loss of correction.
Methods: A prospective, longitudinal study database was retrospectively queried for patients with at least 2 years of follow-up who underwent surgery for adolescent idiopathic scoliosis. Loss of correction was defined (1) clinically, as a >5 degree increase in the inclinometer reading at any time interval after surgery, excluding preoperative values; or (2) radiographically, as an increase of ≥ 10 degrees in the coronal Cobb angle of an instrumented curve. Patients with pseudarthrosis, adding-on, or loss of fixation were excluded.
Results: Ninety-three of 800 patients (11.6%) had loss of deformity correction. Twenty-one patients (2.5%) had >5 degrees of inclinometer change and 76 patients (9.5%) had ≥10 degrees of Cobb change, with a mean loss of 10.85±4.2 degrees. When comparing pedicle screw constructs only, 10% of patients developed loss of correction as opposed to 20% of the patients with hybrid constructs (P=0.036). Factors associated with loss of correction were: age, hybrid construct, magnitude of thoracic curve, thoracic apical translation, and T1 tilt angle.
Conclusion: This study reports a high percentage of loss of coronal correction (12%) after 2 years of follow-up that was not associated with infection, adding-on, or apparent pseudarthrosis. Larger magnitude thoracic curves with apical deviation and hybrid constructs were associated with an increased likelihood of loss of curve correction where use of pedicle screw fixation seems to decrease its incidence.
Level of Evidence: III.