Background: Whether muscle weakness is a cause, or result, of degenerative spondylolisthesis is not currently well understood. Little biomechanical evidence is available to offer an explanation for the mechanism behind exercise therapy. Therefore, the aim of this study is to investigate the effects of back muscle weakness on degenerative spondylolisthesis and to tease out the biomechanical mechanism of exercise therapy.Methods: A nonlinear 3-D finite element model of L3-L5 was constructed. Forces representing global back muscles and global abdominal muscles, follower loads and an upper body weight were applied. The force of the global back muscles was reduced to 75%, 50% and 25% to simulate different degrees of back muscle weakness. An additional boundary condition which represented the loads from other muscles after exercise therapy was set up to keep the spine in a neutral standing position. Shear forces, intradiscal pressure, facet joint forces and von Mises equivalent stresses in the annuli were calculated.Findings: The intervertebral rotations of L3-L4 and L4-L5 were within the range of in vitro experimental data. The calculated intradiscal pressure of L4-L5 for standing was 0.57 MPa, which is similar to previous in vivo data. With the back muscles were reduced to 75%, 50% and 25% force, the shear force moved increasingly in a ventral direction. Due to the additional stabilizing force and moment provided by boundary conditions, the shear force varied less than 15%.Interpretation: Reducing the force of global back muscles might lead to, or aggravate, degenerative spondylolisthesis with forward slipping from biomechanical point of view. Exercise therapy may improve the spinal biomechanical environment. However, the intrinsic correlation between back muscle weakness and degenerative spondylolisthesis needs more clinical in vivo study and biomechanical analysis. (C) 2016 Elsevier Ltd. All rights reserved.