Spinal sagittal alignment goals based on statistical modelling and musculoskeletal simulations

Area of Science:

• Spine biomechanics
• Orthopedic surgery
• Biomedical engineering

Background:

• Current spinal fusion target alignment relies on anatomical criteria and surgical experience, often lacking patient-specific optimization.
• Optimal patient alignment for spinal fusion remains an unknown, impacting surgical outcomes.
• Statistical and musculoskeletal models offer potential for predicting physiological alignments and analyzing biomechanics.

Purpose of the Study:

• To statistically predict patient-specific spinal alignments for fusion surgery.
• To hypothesize that predicted alignments would be biomechanically favorable.
• To evaluate predicted alignments for improved balance and reduced musculoskeletal loads.

Main Methods:

• A statistical model trained on 60 annotated radiographs predicted physiological sagittal alignment.
• Predicted alignments for 11 back pain patients were clinically evaluated for balance against normative ranges.
• Musculoskeletal models simulated loads in upright and flexed postures to compare predicted vs. original alignments.

Main Results:

• The majority of predicted alignments (9/11) met at least two of three balance parameters, compared to original alignments (4/11).
• Predicted alignments significantly reduced overall muscle activity and compressive loads across all levels and postures.
• Significant reductions in shear forces were observed at multiple lumbar levels in both upright and flexed postures.

Conclusions:

• Statistical models can predict spinal alignments that enhance patient balance and decrease musculoskeletal loads.
• Predicted alignments demonstrate biomechanical advantages over traditional alignment criteria.
• Further research is necessary to establish the clinical validity of these predictive models in spinal fusion surgery.