Dynamic forces acting on the lumbar spine during manual handling. Can they be estimated using electromyographic techniques alone?

STUDY DESIGN Compressive loading of the lumbar spine was analyzed using electromyographic, movement analysis, and force-plate techniques. OBJECTIVES To evaluate the inertial forces that cannot be detected by electromyographic techniques alone. SUMMARY OF BACKGROUND DATA Links between back pain and manual labor have stimulated attempts to measure spine compressive loading. However, direct measurements of intradiscal pressure are too invasive, and force plates too cumbersome for use in the workplace. Electromyographic techniques are noninvasive and portable, but ignore certain inertial forces. METHODS Eight men lifted boxes weighing 6.7 and 15.7 kg from the ground, while joint moments acting about L5-S1 were quantified 1) by using a linked-segment model to analyze data from Kistler force plates and a Vicon movement-analysis system, and 2) by measuring the electromyographic activity of the erector spinae muscles, correcting it for contraction speed and comparing it to moment generation during static contractions. The linked-segment model was used to calculate the "axial thrust," defined as the component of the L5-S1 reaction force that acts along the axis of the spine and that is unrelated to trunk muscle activity or static body weight. RESULTS Peak extensor moments predicted by the two techniques were similar and equivalent to spinal compressive forces of 2.9-4.8 kN. The axial thrust "hidden" from the electromyographic technique was negligible during slow lifts, and remained below 4% of peak spinal compression even during fast heavy lifts. Peak axial thrust was proportional to the peak vertical ground reaction (R2 = 0.74). CONCLUSIONS Electromyographic techniques can measure dynamic spinal loading, but additional force-plate data would improve accuracy slightly during lifts requiring a vigorous upward thrust from the legs.