Computational modeling and analysis of thoracolumbar spine fractures in frontal crash reconstruction
Xin Ye,James P. Gaewsky,Derek A. Jones,Logan E. Miller,Joel D. Stitzel,Ashley A. Weaver
DOI: https://doi.org/10.1080/15389588.2018.1498090
2018-10-10
Traffic Injury Prevention
Abstract:OBJECTIVE: This study aimed to reconstruct 11 motor vehicle crashes (6 with thoracolumbar fractures and 5 without thoracolumbar fractures) and analyze the fracture mechanism, fracture predictors, and associated parameters affecting thoracolumbar spine response.METHODS: Eleven frontal crashes were reconstructed with a finite element simplified vehicle model (SVM). The SVM was tuned to each case vehicle and the Total HUman Model for Safety (THUMS) Ver. 4.01 was scaled and positioned in a baseline configuration to mimic the documented precrash driver posture. The event data recorder crash pulse was applied as a boundary condition. For the 6 thoracolumbar fracture cases, 120 simulations to quantify uncertainty and response variation were performed using a Latin hypercube design of experiments (DOE) to vary seat track position, seatback angle, steering column angle, steering column position, and D-ring height. Vertebral loads and bending moments were analyzed, and lumbar spine indices (unadjusted and age-adjusted) were developed to quantify the combined loading effect. Maximum principal strain and stress data were collected in the vertebral cortical and trabecular bone. DOE data were fit to regression models to examine occupant positioning and thoracolumbar response correlations.RESULTS: Of the 11 cases, both the vertebral compression force and bending moment progressively increased from superior to inferior vertebrae. Two thoracic spine fracture cases had higher average compression force and bending moment across all thoracic vertebral levels, compared to 9 cases without thoracic spine fractures (force: 1,200.6 vs. 640.8 N; moment: 13.7 vs. 9.2 Nm). Though there was no apparent difference in bending moment at the L1-L2 vertebrae, lumbar fracture cases exhibited higher vertebral bending moments in L3-L4 (fracture/nonfracture: 45.7 vs. 33.8 Nm). The unadjusted lumbar spine index correctly predicted thoracolumbar fracture occurrence for 9 of the 11 cases (sensitivity = 1.0; specificity = 0.6). The age-adjusted lumbar spine index correctly predicted thoracolumbar fracture occurrence for 10 of the 11 cases (sensitivity = 1.0; specificity = 0.8). The age-adjusted principal stress in the trabecular bone was an excellent indicator of fracture occurrence (sensitivity = 1.0; specificity = 1.0). A rearward seat track position and reclined seatback increased the thoracic spine bending moment by 111-329%. A more reclined seatback increased the lumbar force and bending moment by 16-165% and 67-172%, respectively.CONCLUSIONS: This study provided a computational framework for assessing thoracolumbar fractures and also quantified the effect of precrash driver posture on thoracolumbar response. Results aid in the evaluation of motor vehicle crash-induced vertebral fractures and the understanding of factors contributing to fracture risk.
public, environmental & occupational health,transportation