Neurosurgery Resident Physician Dell Medical School at The University of Texas at Austin
Introduction: This study aimed to explore the relationship between thoracic flexibility, cervical alignment, and clinical outcomes in patients undergoing posterior cervical fusion. It also sought to assess whether advanced predictive models could improve outcome prediction in this patient population.
Methods: A retrospective analysis was conducted on 227 patients who had undergone multi-level posterior cervical fusion. Thoracic flexibility was assessed by measuring the difference in T1 slope between upright X-rays and supine MRI/CT scans. Cervical alignment was quantified using changes in C2-C7 lordosis and sagittal vertical axis (SVA). Clinical outcomes were measured using the Visual Analog Scale (VAS) for pain and the Oswestry Disability Index (ODI) for disability. Analytical techniques included interaction analysis, non-linear regression, clustering (K-means), and predictive modeling using Gradient Boosting Machines (GBM) and Random Forests. Cross-validation and silhouette analysis were employed for model validation.
Results: Interaction analysis showed that thoracic flexibility (Δ T1 slope) and sagittal imbalance (Δ SVA) together moderately impacted disability, but not all outcome variability could be explained. BMI emerged as a key factor, with higher BMI patients experiencing more disability from sagittal imbalance. Non-linear regression revealed a complex relationship between Δ T1 slope and Δ SVA (R² = 0.333), indicating that small changes in thoracic flexibility can disproportionately affect cervical alignment. While these biomechanical factors significantly influenced disability, their impact on pain was minimal. Clustering identified patient subgroups based on flexibility and imbalance, with those showing high flexibility and imbalance experiencing the highest disability and pain. Predictive models (GBM, Random Forest) underperformed, showing poor predictive accuracy, suggesting that additional variables may be needed to improve predictions.
Conclusion : Thoracic flexibility and cervical alignment play a role in disability but have limited predictive power for pain and overall outcomes. BMI and biomechanical interactions exacerbate disability, and incorporating psychosocial factors may improve prediction models. These findings may guide more personalized treatment strategies tailored to specific biomechanical profiles.
.jpg)
