Predictive equations to estimate spinal loads in symmetric lifting tasks

Response surface methodology is used to establish robust and user-friendly predictive equations that relate responses of a complex detailed trunk finite element biomechanical model to its input variables during sagittal symmetric static lifting activities. Four input variables (thorax flexion angle, lumbar/pelvis ratio, load magnitude, and load position) and four model responses (L4–L5 and L5–S1 disc compression and anterior–posterior shear forces) are considered. Full factorial design of experiments accounting for all combinations of input levels is employed. Quadratic predictive equations for the spinal loads at the L4–S1 disc mid-heights are obtained by regression analysis with adequate goodness-of-fit (R2>98%, p<0.05, and low root-mean-squared-error values compared with the range of predicted spine loads). Results indicate that intradiscal pressure values at the L4–L5 disc estimated based on the predictive equations are in close agreement with available in vivo data measured under similar loadings and postures. Combinations of input (posture and loading) variable levels that yield spine loads beyond the tolerance compression limit of 3400 N are identified using contour plots. Ergonomists and bioengineers, faced with the dilemma of using either complex but more accurate models on one hand or less accurate but simple models on the other hand, have thereby easy-to-use predictive equations that quantifies spinal loads and risk of injury under different occupational tasks of interest.