Effects of vertebral shortening during corrective osteotomy for spinal deformity on spinal cord biomechanics and cerebrospinal fluid-anterior spinal artery fluid-structure interaction

Corrective osteotomy for severe spinal deformity requires vertebral shortening, but the safe degree of shortening remains uncertain. This study used fluid-structure interaction finite element analysis to investigate how graded vertebral shortening affects cerebrospinal fluid (CSF) dynamics and anterior spinal artery (ASA) perfusion. A three-dimensional T1-T12 model was reconstructed from computed tomography data obtained from a 16-year-old male with scoliosis. Posterior vertebral column resection at T8/T9 was simulated in four conditions: no shortening and shortening by one-third, one-half, and two-thirds. Coupled CSF-dura and blood-vessel wall models were used to quantify pressure, wall shear stress, flow velocity, structural stress, and displacement. With shortening of up to one-half, CSF wall shear stress (~ 0.04 Pa) and velocity (~ 0.014 m/s) remained stable, whereas ASA luminal pressure and velocity declined gradually together with vessel-wall stress and displacement, indicating preserved hemodynamic compensation and mechanical unloading. At two-thirds shortening, CSF wall shear stress (0.4998 Pa) and velocity (0.16009 m/s) increased by approximately 12-fold and 11-fold, respectively; dural stress increased by approximately 34%, whereas displacement increased by only 1.2%, demonstrating nonlinear stress-displacement decoupling. ASA luminal pressure, velocity, vessel-wall stress, and displacement also rebounded to their highest values. In this patient-specific model, shortening up to one-half was associated with relative stability of CSF dynamics and progressive vascular unloading, whereas two-thirds shortening produced marked nonlinear changes in CSF and ASA-related biomechanical variables. These findings suggest a model-derived biomechanical transition between one-half and two-thirds shortening; however, this transition should not be interpreted as a clinically validated safety threshold.