Kevin P. Granata

An EMG-assisted model of trunk loading during free-dynamic lifting.

One of the continuing challenges in biomechanics has been to assess loading of the spine during dynamic lifting exertions. A model was developed to accurately simulate multi-dimensional spinal loads and trunk moments from measured muscle coactivity and external forces during free-dynamic lifting exertions. Model validity was demonstrated by comparing measured and predicted trunk extension moments. Its purpose was to examine realistic representations of lifting kinetics, kinematics, and dynamic trunk mechanics that may influence spinal loading, and to demonstrate that EMG-assisted modeling techniques can be applied to the analysis of free-dynamic exertions. Spinal loads and trunk moments were predicted from the muscle force vectors and external loads. Muscle tensile forces were determined from the product of normalized EMG data modulated to account for contractile dynamics, muscle cross sectional area, and muscle force per unit cross-sectional area. Model output was physiologically valid, i.e. average predicted muscle force per unit cross-sectional area of 50-65 N cm-2, and accurately predicted measured, dynamic, lifting moments, with an average R2 = 0.81 in the sagittal plane and R2 = 0.76 in the lateral plane. Results indicated that compressive and shear loading increased significantly with exertion load, lifting velocity, and trunk asymmetry.

An EMG-assisted model of loads on the lumbar spine during asymmetric trunk extensions

An EMG-assisted, low-back, lifting model is presented which simulates spinal loading as a function of dynamic, asymmetric, lifting exertions. The purpose of this study has been to develop a model which overcomes the limitations of previous models including static or isokinetic mechanics, inaccurate predictions of muscle coactivity, static interpretation of myoelectric activity, and physiologically unrealistic or variable muscle force per unit area. The present model predicts individual muscle forces from processed EMG data, normalized as a function of trunk angle and asymmetry, and modified to account for muscle length and velocity artifacts. The normalized EMGs are combined with muscle cross-sectional area and intrinsic strength capacity as determined on a per subject basis, to represent tensile force amplitudes. Dynamic internal and external force vectors are employed to predict trunk moments, spinal compression, lateral and anterior shear forces. Data from 20 subjects performing a total of 2160 exertions showed good agreement between predicted and measured values under all trunk angle, asymmetry, velocity, and acceleration conditions. The design represents a significant step toward accurate, fully dynamic modeling of the low-back in multiple dimensions. The benefits of such a model are the insights provided into the effects of motion induced, muscle co-activity on spinal loading in multiple dimensions.

Cost-benefit of muscle cocontraction in protecting against spinal instability

STUDY DESIGN Lifting dynamics and electromyographic activity were evaluated using a biomechanical model of spinal equilibrium and stability to assess cost-benefit effects of antagonistic muscle cocontraction on the risk of stability failure. OBJECTIVES To evaluate whether increased biomechanical stability associated with antagonistic cocontraction was capable of stabilizing the related increase in spinal load. SUMMARY OF BACKGROUND DATA Antagonistic cocontraction contributes to improved spinal stability and increased spinal compression. For cocontraction to be considered beneficial, stability must increase more than spinal load. Otherwise, it may be possible for cocontraction to generate spinal loads that cannot be stabilized. METHODS A biomechanical model was developed to compute spinal load and stability from measured electromyography and motion dynamics. As 10 healthy men performed sagittal lifting tasks, trunk motion, reaction loads, and electromyographic activities of eight trunk muscles were recorded. Spinal load and stability were evaluated as a function of cocontraction and trunk flexion angle. Stability was quantified in terms of the maximum spinal load the system could stabilize. RESULTS Cocontraction was associated with a 12% to 18% increase in spinal compression and a 34% to 64% increase in stability. Spinal load and stability increased with trunk flexion. CONCLUSIONS Despite increases in spinal load that had to be stabilized, the margin between stability and spinal compression increased significantly with cocontraction. Antagonistic cocontraction was found to be most beneficial at low trunk moments typically observed in upright postures. Similarly, empirically measured antagonistic cocontraction was recruited less in high-moment conditions and more in low-moment conditions.

Gender differences in active musculoskeletal stiffness. Part II. Quantification of leg stiffness during functional hopping tasks

Leg stiffness was compared between age-matched males and females during hopping at preferred and controlled frequencies. Stiffness was defined as the linear regression slope between the vertical center of mass (COM) displacement and ground-reaction forces recorded from a force plate during the stance phase of the hopping task. Results demonstrate that subjects modulated the vertical displacement of the COM during ground contact in relation to the square of hopping frequency. This supports the accuracy of the spring-mass oscillator as a representative model of hopping. It also maintained peak vertical ground-reaction load at approximately three times body weight. Leg stiffness values in males (33.9+/-8.7 kN/m) were significantly (p<0.01) greater than in females (26.3+/-6.5 kN/m) at each of three hopping frequencies, 3.0, 2.5 Hz, and a preferred hopping rate. In the spring-mass oscillator model leg stiffness and body mass are related to the frequency of motion. Thus male subjects necessarily recruited greater leg stiffness to drive their heavier body mass at the same frequency as the lighter female subjects during the controlled frequency trials. However, in the preferred hopping condition the stiffness was not constrained by the task because frequency was self-selected. Nonetheless, both male and female subjects hopped at statistically similar preferred frequencies (2.34+/-0.22 Hz), therefore, the females continued to demonstrate less leg stiffness. Recognizing the active muscle stiffness contributes to biomechanical stability as well as leg stiffness, these results may provide insight into the gender bias in risk of musculoskeletal knee injury.

Gender differences in active musculoskeletal stiffness. Part I.: Quantification in controlled measurements of knee joint dynamics

Active females demonstrate increased risk for musculoskeletal injuries relative to equivalently-trained males. Although gender differences in factors such as passive laxity, skeletal geometry and kinematics have been examined, the effect of gender on active muscle stiffness has not been reported. Stiffness of the active quadriceps and hamstrings musculature were recorded during isometric knee flexion and extension exertions from twelve male and eleven female subjects. A second-order biomechanical model of joint dynamics was used to quantify stiffness from the transient motion response to an angular perturbation of the lower-leg. Female subjects demonstrated reduced active stiffness relative to male subjects at all torque levels, with levels 56-73% of the males. Effective stiffness increased linearly with the torque load, with stiffness increasing at a rate of 3.3 Nm/rad per unit of knee moment in knee flexion exertions (hamstrings) and 6.6 Nm/rad per unit of knee moment extension exertions (quadriceps). To account for gender differences in applied moment associated with leg mass, regressions analyses were completed that demonstrated a gender difference in the slope of stiffness-versus-knee moment relation. Further research is necessary to identify the cause of the observed biomechanical difference and implications for controlling injury.

Female and male trunk geometry: size and prediction of the spine loading trunk muscles derived from MRI.

OBJECTIVE Develop a gender specific database of trunk muscle cross-sectional areas across multiple levels of the thoracic and lumbar spine and develop prediction equations for the physiological cross-sectional area as a function of gender and anthropometry. DESIGN This study quantified trunk muscle cross-sectional areas of male and female spine loading muscles. BACKGROUND There is a lack of comprehensive data regarding the female spine loading muscle size. Although biomechanical models often assume females are the same as males, little is known regarding gender differences in terms of trunk muscle areas and no data exist regarding the prediction of trunk muscle physiological cross-sectional areas from commonly used external anthropometric measures. METHODS Magnetic resonance imaging scans through the vertebral bodies from T(8) through S(1) were performed on 20 females and 10 males. Muscle fiber angle corrected cross-sectional areas were recorded at each vertebral level. Linear regression techniques taking into account anthropometric measures were utilized to develop prediction equations for the physiological cross-sectional area for each muscle of interest, as well as tests for differences in cross-sectional areas due to gender and side of the body. RESULTS Significant gender differences were observed for the prediction of the erector spinae, internal and external obliques, psoas major and quadratus lumborum physiological cross-sectional areas. Anthropometric measures about the xyphoid process and combinations of height and weight resulted in better predictions of cross-sectional areas than when using traditional anthropometry. CONCLUSIONS This study demonstrates that the trunk muscle geometry of females and males are different, and that these differences should be considered in the development of biomechanical models of the torso. Relevance. The prediction of physiological cross-sectional areas from external anthropometric measures provide gender specific equations to assist in estimation of forces of muscles which load the spine for biomechanical purposes.

The Influence of Trunk Muscle Coactivity on Dynamic Spinal Loads

Study Design Measured trunk muscle activity was employed in a biomechanical model to determine the influence of including or neglecting muscle coactivity on predicted spinal loads. Objectives The purpose of this investigation was to examine the influence of muscle coactivity on spinal load. Summary of Background Data Electromyographic patterns in the trunk musculature have demonstrated significant levels of cocontraction during lifting exertions. Biomechanical analyses of musculoskeletal loading are often mathematically constrained from including muscle coactivity. Models that attempt to include coactive behavior are complex and difficult to implement. Methods Electromyographic data were collected from five trunk muscle pairs while subjects performed dynamic lifting excertions. A validated, electromyographically assisted biomechanical model was used to compute relative muscle force, lifting moment, and spinal load. Results were generated and compared from analyses that included from one to five simultaneously active muscle pairs. Results Trunk extensor muscles generate lifting moments as much as 47% greater than the applied lifting moment to offset flexor antagonism. Analyses that neglect muscle coactivity during dynamic lifting exertions may underestimate spinal compression by as much as 45% and shear forces by as much as 70%. Conclusions The level of coactive spinal loading is significantly influenced by the weight of the lifted load as well as trunk extension velocity. Muscle coactivity significantly influences the modeled load in the lumbar spine during lifting exertions as should be considered if an accurate measure of spinal loading of desired.

A biomechanical assessment and model of axial twisting in the thoracolumbar spine.

Study Design Measured trunk kinematics, applied moments, and trunk muscle activities were employed in a biomechanical model to determine load experiences by the spine during dynamic torsional exertions. Objectives The purpose of this investigation was to examine the influence of dynamic twisting parameters on spinal load. Summary of Background Data Axial twisting of the torso has been identified as a significant risk factor for occupationally related low back disorders. However, previous studies have had difficulty describing how twisting is accomplished biomechanically, or how the spine is loaded during twisting motions. Methods Electromyograph activity of 10 trunk muscles was monitored while 12 subjects performed twisting exertions under various conditions of force, velocity, position, and direction. An electromyograph-assisted biomechanical model was developed to interpret the effects of those twisting parameters on spine loading. Results Significant flexion-extension and lateral moments were generated during the twisting exertions. Muscle coactivity associated with twisting exertions was significantly greater than that associated with lifting exertions. Employing electromyograph data to represent muscle coactivity, the model accurately predicted trunk moments and hence was assumed to reasonably reflect spine loading. Conclusions Under the conditions tested, the results indicated that relative spinal compression during dynamic twisting exertions was twice that of static exertions. Spine loading also varied as a function of whether the trunk was twisted to the left or right and according to the direction of applied torsion—i.e., clockwise or counterclockwise. The results may help explain, biomechanically, why epidemiologic findings have repeatedly identified twisting as a risk factor for low back disorder.

Trunk posture and spinal stability.

OBJECTIVE The influence of trunk posture on musculoskeletal stability of the spine was investigated. DESIGN A biomechanical model was developed to evaluate the influence of posture on spinal stability. Model performance was assessed by comparing predicted muscle recruitment patterns with measured EMG activity from the trunk muscles during static lifting exertions. METHOD An inverted double-pendulum model of the spine controlled by 12 muscle equivalents of the trunk was implemented to determine spinal load and stability. Model input included trunk posture and lifted mass, output included muscle recruitment patterns necessary to achieve stability of the spine and spinal load. EMG activity recorded from the trunk muscles of 10 subjects were recorded during static exertions in various trunk flexion and asymmetric postures to compare with model output. Stable spinal load was examined as a function of trunk flexion and asymmetry during the lifting exertions. RESULTS Antagonistic co-contraction was necessary to achieve spinal stability, particularly in upright postures. Stable spinal load was increased in asymmetric postures as a result of antagonistic muscle recruitment, suggesting greater neuromuscular control is necessary to maintain stability in asymmetric lifting postures. As trunk flexion angle increased, stability improved but spinal load was greater. CONCLUSIONS Results illustrate that muscle recruitment patterns are more accurately explained by stability than by equilibrium alone. Spinal stability is influenced by posture. Specifically, control of spinal stability is reduced in asymmetric postures associated with low-back disorder risk. RELEVANCE Traditional assessment of low-back disorder risk have focussed on spinal loading. Results illustrate that postural risk factors for low-back pain may be partially attributable to stability considerations.

Response of trunk muscle coactivation to changes in spinal stability

The goal of this effort was to assess the neuromuscular response to changes in spinal stability. Biomechanical models suggest that antagonistic co-contraction may be related to stability constraints during lifting exertions. A two-dimensional biomechanical model of spinal equilibrium and stability was developed to predict trunk muscle co-contraction as a function of lifting height and external load. The model predicted antagonistic co-contraction must increase with potential energy of the system even when the external moment was maintained at a constant value. Predicted trends were compared with measured electromyographic (EMG) data recorded during static trunk extension exertions wherein subjects held weighted barbells at specific horizontal and vertical locations relative to the lumbo-sacral spine junction. The task was designed to assure the applied moment was identical during each height condition, thereby changing potential energy without influencing moment. Measured EMG activity in the trunk flexors increased with height of the external load as predicted by the model. Gender difference in spinal stability were also noted. Results empirically demonstrate that the neuromuscular system responds to changes in spinal stability and provide insight into the recruitment of trunk muscle activity.