LAUSR

Study Design. A retrospective study of consecutive patients undergoing surgery for adult spinal deformity (ASD). Objective. The aim of this study was to classify postoperative ASD patients by sagittal compensation and characterize the features of each group. Summary of Background Data. Sagittal compensatory mechanisms to keep the erect position would function in not only nonoperative ASD patients but also in postoperative patients. However, details of sagittal compensatory mechanisms after surgery have been unclear, because the majority of previous studies examined the compensatory mechanisms in nonoperative patients. Methods. A total of 73 postoperative ASD patients were recruited. Spinopelvic parameters and Oswestry Disability Index (ODI) were measured before surgery, at 1-month and 2-year follow-up. The changes in parameters (&Dgr;) between postoperative 1-month and 2-year were also evaluated. Cluster analysis based on pelvic tilt (PT) and sagittal vertical axis (SVA) at 2-year follow-up was performed, and then the parameters were compared among clusters. Results. Cluster analysis identified three clusters: the physiological group (n = 38, PT = 19.6°, SVA = 17.2 mm), the compensated group (n = 23, PT = 34.3°, SVA = 45.6 mm), and the decompensated group (n = 12, PT = 36.6°, SVA = 118.8 mm). Comparisons of the parameters among the groups showed that the physiological group had smaller pelvic incidence minus lumbar lordosis (PI-LL) and better ODI. The compensated group had significantly larger &Dgr; thoracic kyphosis (TK), &Dgr; proximal junctional angle (PJA), and &Dgr;PT than the physiological group. The decompensated group had significantly higher PI and higher age. Conclusion. Three groups were defined in terms of postoperative compensatory mechanisms. Patients in the physiological group had little postoperative alignment change and good ODI value. The compensated group had pelvic retroversion to keep SVA in the physiological range, but the compensation in the thoracic spine was unlikely to work. High PI and advanced age were characteristics of the decompensated group. Level of Evidence: 4