Stabilization of the lumbar spine by spinal muscle forces producing compressive follower loads: 3‐dimensional computational study

Abstract

Axial compressive loads whose direction changes along the spinal curvature (so called compressive follower loads (CFLs)) was postulated as a normal physiological load in the lumbar spine in the literature. Computational analyses were conducted in this study using finite element and optimization models of the spinal system incorporating 244 fascicles of back muscles. It was feasible to find optimum solutions for spinal muscle forces generating CFLs in the lumbar spine in 3-D postures of neutral standing, flexion 40°, extension 10°, axial rotation 10°, or lateral bending 30°. FE analyses demonstrated that the lumbar spine can be in a stable condition not under all CFL generating muscle forces but under those producing CFLs along a curve parallel to the spinal curvature located in the vicinity of the base spinal curve constructed by connecting the geometrical centers of the vertebral bodies. It was also possible to estimate the stable range of the relative location of such CFL curve to the base spinal curve. These results suggest that the lumbar spine in various 3-D postures can be stabilized by spinal muscles that generate CFLs in the spine, which at least in part supports the hypothesis of CFLs as a physiological load in the lumbar spine. © 2018 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res.

Computational modelling of the scoliotic spine: A literature review

“… 50 Dreischarf et al 52 simulated realistic motions by optimising the magnitude to minimise the intervertebral rotations, in another study the magnitude was optimised to minimise the IVJ load. 51 In a number of studies the follower load has been assumed to be task independent. 16 , 28 , 53 According to Azari et al 54 this is not the case.…”

“… 29 , 57 , 58 One approach is to combine a FEM and an optimization method to calculate muscle forces. 29 , 39 , 45 , 51 In other cases muscles are modelled simply using tension elements 57 while a more sophisticated model has also accounted for the muscle geometry and architecture. 58 Inclusion of the muscles and the intra‐abdominal pressure decreased the predicted stress in the IVDs.…”

“…Other studies have modelled the individual components of the IVJ. Linear 32 and non‐linear models 26 , 27 , 28 , 34 , 35 , 37 , 49 , 51 have been used for the IVD, often modelling the nucleus pulposus and the annulus fibrosus separately. Non‐linear hyperelastic equations described the RoM much better than linear equations.…”

“…Non-linear hyperelastic equations described the RoM much better than linear equations. 59 The ligaments have been modelled as both linear 32,51 and non-linear. [26][27][28]35,37,49 Naserkhaki et al 34 found a piecewise linear model from an in vivo dataset best predicted the mean in vitro rotations, and higher non-linearity caused greater movement of the instantaneous CoR .…”

Re: Foresto T, Song I, Kim BS, Lim TH. 2018. Stabilization of the lumbar spine by spinal muscle forces producing compressive follower loads: 3‐dimensional computational study

“…We disagree with the conclusions of the recent paper entitled “Stabilization of the lumbar spine by spinal muscle forces producing compressive follower loads: 3‐dimensional computational study.” In this paper the authors use an analytical model of the lumbar spine and its muscles to estimate the activation of the musculature consistent with several constraints including that the path of net forces acting between adjacent vertebrae be close to the vector joining the centers of the vertebrae. This configuration (alignment) of the intervertebral forces has been described as a “follower load.” If compressive forces do act “off‐center” they will generate moments, that in turn will produce potentially large intersegmental rotations; hence motion of the trunk.…”