Poster Presentation 50th International Society for the Study of the Lumbar Spine Annual Meeting 2024

Introduction: Revision rates following spinal realignment surgery for adult spinal deformity can range from 20-30%. To assess post-operative outcomes, health-related quality of life (HRQOL) questionnaire assessments are heavily utilized and are measuring a patient’s health experience. It is unclear how post-operative recovery patterns based on patient-reported outcomes may correspond to the correction in spinal alignment and risk for revision surgery. Past studies have shown correlations between individual patient outcomes and sagittal alignment parameters at a single timepoint [1,2]. However, it is unclear how longitudinal recovery trends in patient-reported outcomes relate to changes in sagittal alignment. The purpose of this study is to identify patient-specific recovery patterns based on HRQOL outcomes and understand how they relate to changes in spinal alignment following surgery. We hypothesize that HRQOL-based recovery response patterns spanning 1-year following surgery will relate to sagittal alignment measurements and may be used to understand if patients are at risk for needing a revision.

Methods: Patient recovery was assessed using the Scoliosis Research Society Composite Score (SR22), which represents function, pain, mental health, self-image, and satisfaction with treatment preoperatively, at six weeks, three months, and one year. Spinal alignment was assessed by measuring SVA, lumbar lordosis (LL), thoracic kyphosis (TK), and mismatch between pelvic incidence and lumbar lordosis (PI-LL) on preoperative and postoperative full-length standing radiographs. Latent class mixed models were used to assess SRS score over time, to identify unique recovery groups. Associations between sagittal alignment measurements and recovery trajectory groups were assessed using ANOVA. 

Results: 170 patients had complete data and were eligible for inclusion. We identified three groups with distinct recovery trajectories (Figure 1): Gradual responders (GR; 83%), Super responders (SR; 10%), and Non-responders (NR; 7%). The response groups did not associate with any of the pre-operative or post-operative sagittal alignment factors, except there was a trend for pre-operative LL (p=0.06) being significantly larger for the NR group (52.7±13.6) compared to the GR (38.8±18.4) and SR (35.7±21.7) groups. There was however an association between magnitude of correction and response group for TK (GR: 8.3±12.8; SR: 2.0±17.1; NR: 2.3±16.4; p=0.04; Figure 1) and a trend with LL (GR: 15.3±16.9; SR: 14.5±19.7; NR: 5.2±13.1; p=0.08). We did not find significant associations with changes in SVA or PI-LL with recovery response group, although SVA appears to be distinct between responder groups (Figure 1) but is likely underpowered.  

Conclusion: Using patient-reported outcomes for assessing a patient’s health experience, we identified three distinct recovery response patterns following sagittal realignment surgery. The majority of which (83%) fell into the Gradual responder group, which also had the most positive change in TK and LL correction with surgery. The Non-responder group (7%) has a comparable health status at baseline, but less post-operative change in TK and LL. It remains unclear how well recovery patterns in post-operative outcomes based on HRQOL can relate to post-operative surgical outcomes and risk for revision surgery. Future work will use electronic health record data to predict recovery response patterns before surgery and associate recovery response patterns with revision surgeries.