BackgroundIndividual study results have demonstrated unclear relationships between neurocompressive disorders and paraspinal muscle morphology. This systematic review aimed to synthesize current evidence regarding the relationship lumbar neurocompressive disorders may have with lumbar paraspinal muscle morphology.MethodsSearches were conducted in seven databases from inception through October 2017. Observational studies with control or comparison groups comparing herniations, facet degeneration, or canal stenosis to changes in imaging or biopsy-identified lumbar paraspinal muscle morphology were included. Data extraction and risk of bias assessment were performed by review author pairs independent of one another. Morphological differences between individuals with and without neurocompressive disorders were compared qualitatively, and where possible, standardised mean differences were obtained.ResultsTwenty-eight studies were included. Lumbar multifidus fiber diameter was smaller on the side of and below herniation for type I [SMD: −0.40 (95% CI = −0.70, −0.09) and type II fibers [SMD: −0.38 (95% CI = −0.69, −0.06)] compared to the unaffected side. The distribution of type I fibers was greater on the herniation side [SMD: 0.43 (95% CI = 0.03, 0.82)]. Qualitatively, two studies assessing small angular fiber frequency and fiber type groupings demonstrated increases in these parameters below the herniation level. For diagnostic imaging meta-analyses, there were no consistent differences across the various assessment types for any paraspinal muscle groups when patients with herniation served as their own control. However, qualitative synthesis of between-group comparisons reported greater multifidus and erector spinae muscle atrophy or fat infiltration among patients with disc herniation and radiculopathy in four of six studies, and increased fatty infiltration in paraspinal muscles with higher grades of facet joint degeneration in four of five studies. Conflicting outcomes and variations in study methodology precluded a clear conclusion for canal stenosis.ConclusionsBased on mixed levels of risk of bias data, in patients with chronic radiculopathy, disc herniation and severe facet degeneration were associated with altered paraspinal muscle morphology at or below the pathology level. As the variability of study quality and heterogeneous approaches utilized to assess muscle morphology challenged comparison across studies, we provide recommendations to promote uniform measurement techniques for future studies.Trial registrationPROSPERO 2015: CRD42015012985Electronic supplementary materialThe online version of this article (10.1186/s12891-018-2266-5) contains supplementary material, which is available to authorized users.