A Comparative Analysis on Risk Assessment Tools for Elective Spinal Deformity Surgery

Introduction: With a growing number of spinal deformity surgeries and complication rates ranging from 10-80%, development of robust analytic tools for risk assessment could help guide complex medical decision making and enhance value-based care. Existing tools use data points such as comorbid conditions, procedural complexity, and anatomic factors to calculate risk. These tools have even been shown to outperform clinician judgment. However, most models assess specific parameters without external validation, limiting their generalizability. This study evaluates 8 risk assessment tools to identify common and unique predictive factors, laying a foundation for a comprehensive model.

Methods: Eight risk assessment tools were analyzed. Key factors, such as age, diabetes, hypertension, and unique spinal parameters, were compared to identify commonalities and gaps in risk predictions. The study also evaluated each tool’s reliance on internal validation and specific complication types to identify limitations in current models.

Results: Age was the only universally included factor, emphasizing its general influence on outcomes. Diabetes and hypertension both appeared five tools, associated with infection and cardiovascular risks. BMI and smoking were found in 2 and 3 tools respectively, and were predictive of infection, delayed fusion, and mechanical complications. Unique parameters varied by model focus: the GAP Score prioritized biomechanical alignment, while the Seattle-Spine-Score emphasized comorbidities and others emphasized procedural complexity or invasiveness. Fewer than half of the tools had undergone external validation. Overall predictive accuracy ranged from AUC 0.64 to 0.92.

Conclusion : While each tool offers valuable insights into specific complication risks, the lack of an all-encompassing model limits predictive accuracy. A universal tool incorporating both common systemic and spine-specific parameters could enhance risk stratification and support value-based care. This study provides a foundation for developing a multi-domain, externally validated model to enhance risk assessment in spinal deformity surgery.