Describing Novel Spinal Alignment Metrics: The Global Spinal Shear Vector Angle and Magnitude

Introduction: Proximal junctional kyphosis (PJK) occurs in 17%-46% of patients following adult spinal deformity surgery and is associated with worsening quality of life, neurological deficit, and reoperation. Age, bone density, sagittal vertical axis (SVA), and weight have all been linked to PJK risk; however, prediction and prevention of PJK remains challenging. Current deformity correction surgery relies on scalar values including SVA, pelvic incidence, and lumbar lordosis to restore alignment while weight vector metrics may better capture variations in patient anatomy and surgical response.

Methods: We retrospectively identified 20 non-instrumented patients with scoliosis X-Rays in this proof-of-concept study. Patients were stratified by SVA into 4 groups of 5: < 0cm, 0-5cm, 5-10cm, and >10cm. Slope angles at every level were combined with patients’ weights to calculate decomposed weight vectors into their normal and shear components at each spinal level. Global spinal shear vector magnitude (GSV-M) in newtons and angle (GSV-A) in degrees were calculated by summing the individual vectors from every spinal level from axis to sacrum.

Results: The mean (standard deviation) of the GSV-M and GSV-A were 103.3 (55.5) newton and 65.1 (28.3) degrees, respectively. The correlation coefficients were -0.9 between GSV-A and SVA, 0.77 between GSV-M and SVA, -0.68 between GSV-A and GSV-M, 0.47 between GSM-M and weight. The normalized variance is 0.087 for SVA and 0.13 for GSV-A.

Conclusion : GSV-A closely approximated SVA; however, greater variance in GSV-A suggests it may capture greater patient variability compared to SVA alone. GSV-M weakly correlated with both GSV-A and SVA suggesting potentially distinct contributions to spinal alignment. GSV-M moderately correlated with weight scalars suggesting additional contribution of local vertebral slopes and weight vectors to GSV-M. GSV-A and GSV-M can provide more detailed information on “optimal” spinal alignment at each level using two values and help in quantifying PJK risk.