Lumbar muscle volume in postmenopausal women with osteoporotic compression fractures: quantitative measurement using MRI

Objective: To investigate the relationship between paraspinal and psoas muscle volumes and acute osteoporotic or low-bone-mass compression fractures of the lumbar spine in postmenopausal women. Methods: Patient data were retrieved retrospectively for postmenopausal women with L-spine magnetic resonance imaging (MRI) and dual-energy X-ray absorptiometry showing osteoporosis/low bone mass. Group 1 comprised eight women aged 60–80 years with MRI showing a single acute compression fracture. The age-matched group 2a (N = 12) and younger group 2b (N = 12) comprised of women whose MRIs showed no fractures. Cross-sectional MRIs of the paraspinal and psoas muscles and intramuscular fat volume for each muscle group were measured. Operator repeatability and reproducibility were obtained. Results: Group 1 showed significantly smaller lean muscle volume for all muscle groups at L5/S1. Intramuscular fat volume was also smaller in most muscle groups in group 1, though only reaching statistical significance at variable muscle groups and levels. Measurements show both good intrarater repeatability and interrater reproducibility of lean muscle volume estimations (intraclass correlation coefficient (ICC), 0.999 for rater A and 0.997 for rater B; Cronbach’s alpha 0.995) and intramuscular fat volume estimations (ICC, 0.995 for rater A and 0.982 for rater B; Cronbach’s alpha was 0.981). Conclusions: This study provides the first quantitative evidence that compression fractures in postmenopausal women with underlying osteoporosis/low bone mass are associated with less paraspinal and psoas muscle volumes. Further longitudinal studies with larger cohorts are needed to verify this relationship. Key Points: • The risk of osteoporotic compression fractures is higher in older women with smaller paraspinal muscle volume. • Older women show smaller paraspinal muscle volume and more intramuscular fat compared to younger controls.