A biomechanical model of the lumbar spine during upright isometric flexion, extension, and lateral bending

Task-specific and subject-specific lumbar trunk muscle function, muscle geometry, and vertebral density data were collected from 16 men. A biomechanical model was used to determine muscle strength and the compressive forces acting on the lumbar spine.

To develop an anatomic biomechanical model of the low back that could be used to derive task-specific muscle function parameters and to predict compressive forces acting on the low back. Several model-specific constraints were examined, including the notion of bilateral trunk muscle anatomic symmetry, the influence of muscle lines of action, and the use of density-derived vertebral strength for model validation.

Clinical and basic science investigators are currently using a battery of diverse biomechanical techniques to evaluate trunk muscle strength. Noteworthy is the large variability in muscle function parameters reported for different subjects and for different tasks. This information is used to calculate forces and moments acting on the low back, but limited data exist concerning the assessment of subject-specific, multiaxis, isometric trunk muscle functions.

A trunk dynamometer was used to measure maximum upright, isometric trunk moments in the sagittal (extension, flexion) and coronal (lateral flexion) planes. Task- and subject-specific trunk muscle strength or "gain" was determined from the measured trunk moments and magnetic resonance image-based muscle cross-sectional geometry. Model-predicted compressive forces obtained using muscle force and body force equilibrium equations were compared with density-derived estimates of compressive strength.

Individual task-specific muscle gain values differed significantly between subjects and between each of the tasks they performed (extension > flexion > lateral flexion). Significant differences were found between left side and right side muscle areas, and the lines of action of the muscles deviated significantly from the vertical plane. Model-predicted lumbar compressive forces were 38% (lateral flexion) to 73% (extension) lower that the L3 vertebral compressive strength estimated from vertebral density.

The present study suggests that biomechanical models of the low back should be based on task-specific and subject-specific muscle function and precise geometry. Vertebral strength estimates based upon vertebral density appear to be useful for validation of model force predictions.