SPINE Volume 25, Number 15, pp 1957–1964 ©2000, Lippincott Williams & Wilkins, Inc. Decrease in Trunk Muscular Response to Perturbation With Preactivation of Lumbar Spinal Musculature Ian A.F. Stokes, PhD,* Mack Gardner-Morse, MS,* Sharon M. Henry, PT, PhD,† and Gary J. Badger, MS‡ Study Design. An experimental study of healthy sub- jects’ trunk muscle responses to force perturbations at differing angles and steady state efforts. Objectives. To determine whether increased preacti- vation of muscles was associated with decreased likeli- hood of muscular activation in response to a transient force perturbation. Summary of Background Data. Trunk stability (ability to return to equilibrium position after a perturbation) re- quires the stiffness of appropriately activated muscles to prevent buckling and consequent “self-injury.” Therefore, greater trunk muscle preactivation might decrease the likelihood of reflex muscle responses to small perturba- tions. Methods. Each of 13 subjects stood in an apparatus with the pelvis immobilized. A harness around the thorax provided a preload and a force perturbation by a horizon- tal cable and a movable pulley attached to one of five anchorage points on a wall track surrounding the subject at angles of 0°, 45°, 90°, 135°, and 180° to the forward direction. Subjects first equilibrated with a preload effort of nominally 20% or 40% of their maximum extension effort. Then a single full sine-wave force perturbation pulse of nominal amplitude, 7.5% or 15% of maximum effort, duration 80 milliseconds or 300 milliseconds, was applied at a random time, with three repeated trials of each test condition. The applied force (via a load cell) and the electromyographic activity of six right and left pairs of trunk muscles were recorded. Muscle responses were detected by two methods. 1) Shewhart method: electro- myographic signal greater than “baseline” values by more than three standard deviations, and 2) Mean Elec- tromyographic Difference method: mean electromyo- graphic signal in a time window 25 to 150 milliseconds after the force perturbation greater than that in a 25- to 150-millisecond window before the perturbation. Results. Lower preload efforts were associated with more muscle responses (overall mean response detection rate = 33% at low preload and 25% at high preload). Using the Shewhart method, there were significant differ- ences by effort (P  0.05) for all abdominal muscles and for all left dorsal muscles except multifidus. Using the Mean Electromyographic Difference method, there were significant differences by effort (P  0.05) for the same dorsal muscles, but only for one of the abdominal muscles. Conclusions. Findings are consistent with the hypoth- esis that the spine can be stabilized by the stiffness of activated muscles, obviating the need for active muscle responses to perturbations. [Key words: spine stability, muscle stiffness, perturbation, human subjects, electro- myography] Spine 2000;1957–1964 The vertebrae of the lumbar spine are like a series of inverted pendulums, producing a ligamentous spine that is inherently unstable. Stability in this context is defined as the ability of a system to return to its equilibrium position after a small perturbation. The spine must be stabilized by the stiffness of the muscles and motion seg- ments to prevent buckling. Otherwise, a sudden exces- sive displacement could occur and result in tissue injury. It has been shown analytically that muscle stiffness, which increases with intensity of muscle activation, can prevent lumbar spine buckling that would occur other- wise in subjects under loaded conditions or when expe- riencing a perturbation. 2,5,7–11 Therefore, the patterns of human trunk muscle recruitment not only must provide static equilibrium and appropriate response to changes in loading and displacement perturbations, but also must provide sufficient stiffness to ensure stability of the ver- tebral column. 1,2,6,7,10,14,16,18,19,24,25,27,30 Coactivation of antagonistic muscles is a part of a strategy that can increase the muscular stiffness and hence stability, but at the cost of increased spinal loads. 11,12,20,21 The restoration of equilibrium after a perturbation can be achieved by active adjustment of muscle tensions, but with inherent neuromuscular delays. 28,29 Alterna- tively, small perturbations might be accommodated without such active responses, provided there is sufficient muscular stiffness and damping in the trunk. The present study is concerned with the second mechanism. The bi- omechanics of such passive stabilization that requires no active central nervous system-mediated adjustment of the preset muscle activation or stiffness after a perturba- tion was demonstrated theoretically by Bergmark, 2 and has been further explored analytically. 5,7–11 This mech- anism of trunk stabilization is difficult to study experi- mentally, and for practical and ethical reasons experi- mental investigations of stability in human subjects must focus on the strategies that are used to prevent spinal buckling or other instability events. The present study was designed to investigate spinal stability by recording whether trunk muscles were re- cruited in response to a transient perturbation, with dif- ferent magnitudes of preloading. Because muscle stiffness increases with activation, it was expected that the trunk would be stiffer under more heavily loaded conditions; From the Departments of *Orthopaedics and Rehabilitation, †Physical Therapy, and ‡Medical Biostatistics, University of Vermont, Burling- ton, Vermont. Supported by National Institutes of Health R01 AR 44119. Acknowledgement date: June 22, 1999. First revision date: September 7, 1999. Acceptance date: November 23, 1999. Device status category: 1. Conflict of interest category: 14. 1957