GLP-1 Agonist Use Is Not Associated With a Reduction of Paraspinal Muscle Quality in Diabetic Patients With Low Back Pain: A Retrospective Cohort Study

Background: The body-composition-modifying effects of glucagon-like peptide-1 receptor agonists (GLP-1 RAs) remain incompletely characterized, particularly by direct measures of skeletal muscle beyond dual-energy X-ray absorptiometry. This knowledge gap puts vulnerable populations at risk. Purpose: We sought to answer the following question: Is GLP-1 RA use associated with changes in muscle quality as defined by total muscle area, functional cross-sectional area (fCSA), and fatty infiltration (FI), and measured by serial magnetic resonance imaging (MRIs) in the paraspinal musculature of diabetic patients with low back pain? Methods: We conducted a single-center, retrospective cohort study including patients with diabetes and low back pain with at least 2 imaging studies taken at least 3 months apart between 2014 and 2024. Total muscle area, fCSA, and FI were calculated for each patient in the bilateral psoas and posterior paraspinal muscles on T2-weighted axial MRIs. Normalized muscle change per year was calculated between scans for each patient (the difference between muscle measurement on the most recent and the baseline MRI, divided by the length of follow-up). Multivariable linear regression models adjusted for age, baseline body mass index, and sex examined the association of GLP-1 use and muscle changes per year. Results: We examined data on 38 patients on a GLP-1 RA (61% male, median age 67 years) and 128 patients on any other diabetes medication (54% male, median age 71 years). The average duration of treatment for the GLP-1 RA cohort was 21.7 ± 10.7 months, with baseline imaging predating treatment by an average of 10.4 ± 10.1 months. We observed no significant longitudinal differences in estimates of FI or fCSA in the psoas, erector spinae, or multifidus in GLP-1 users compared to controls. Conclusion: This retrospective single-institution study observed no statistically significant differences in changes in paraspinal muscle quality in patients with diabetes and low back pain taking a GLP-1 RA compared to controls. However, this preliminary study may have been underpowered to detect small to moderate effects. Further adequately powered, long-term studies are essential to establish the effects of GLP-1 RAs on muscle quality, especially research using standardized imaging metrics of body composition and muscle quality. Level of Evidence: Level III, retrospective cohort study.