Lumbar paraspinal muscle morphology is associated with spinal degeneration in patients with lumbar spinal stenosis.

BACKGROUND CONTEXT: Lumbar spinal stenosis (LSS) has been reported to induce changes in paraspinal muscle morphology, but objective physical function and degenerative spine conditions are rarely assessed.

PURPOSE: To identify factors associated with paraspinal muscle morphology using objective physical and degenerative spine assessments in patients with LSS.

STUDY DESIGN/SETTING: Cross-sectional design.

PATIENT SAMPLE: Seventy patients with neurogenic claudication caused by LSS, receiving outpatient physical therapy.

OUTCOME MEASURES: Cross-sectional area (CSA) and functional CSA (FCSA) of the multifidus, erector spinae, and psoas muscles, the severity of stenosis, disc degeneration, and end plate abnormalities were evaluated by magnetic resonance imaging, as well as sagittal spinopelvic alignment by X-ray. Objective physical assessments included pedometry and claudication distance. Patient-reported outcomes included the numerical rating scale of low back pain, leg pain, and leg numbness, and the Zurich Claudication Questionnaire.

METHODS: To assess the impact of LSS on paraspinal muscles, FCSA and FCSA/CSA were compared between the dominant and nondominant sides based on the patients' neurogenic symptoms, and multivariable regression analyses adjusted for age, sex, height, and weight were performed; P < 0.05 was considered significant.

RESULTS: Seventy patients were analyzed. At one level below the maximum stenotic level, erector spinae FCSA on the dominant side was significantly lower than that on the nondominant side. In the multivariable regression analyses, at one level below the symptomatic level, disc degeneration, endplate abnormalities, and lumbar spinopelvic alignment, such as decreased lumbar lordosis and increased pelvic tilt, were negatively associated with multifidus FCSA and FCSA/CSA ratio. A significant association was observed between dural sac CSA and erector spinae FCSA. Throughout L1/2 to L5/S, disc degeneration, endplate abnormalities, and lumbar spinopelvic alignment were negatively associated with multifidus and erector spinae FCSA or FCSA/CSA.

CONCLUSIONS: Lumbar paraspinal muscle asymmetry caused by LSS was observed only in erector spinae. Disc degeneration, endplate abnormalities, and lumbar spinopelvic alignment, rather than spinal stenosis and LSS symptoms, were more associated with paraspinal muscle atrophy or fat infiltration.