Related Experiment Video
Updated: Sep 4, 2026
Diffusion Tensor Magnetic Resonance Imaging in Chronic Spinal Cord Compression
Published on: May 7, 2019
MRI-defined erector spinae fatty infiltration predicts progression of sagittal spinal malalignment: a
Keiji Nagata1, Hiroshi Hashizume1, Shizumasa Murata1
1Department of Orthopaedic Surgery, Wakayama Medical University, Wakayama City, Wakayama, Japan.
Background Context:
Sagittal malalignment in adult spinal deformity (ASD) is associated with disability and reduced quality of life. Although paraspinal muscle degeneration has been implicated in spinal alignment changes, longitudinal predictors of ASD progression remain unclear.
Purpose:
To determine whether MRI-defined erector spinae fatty infiltration and pelvic tilt predict sagittal malalignment progression in a population-based cohort.
Study Design/Setting:
Prospective longitudinal cohort study.
Patient Sample:
A total of 533 community-dwelling adults from the Wakayama Spine Study with baseline whole-spine MRI and 6-year radiographic follow-up.
Outcome Measures:
Progression of sagittal malalignment defined as an increase in C7 sagittal vertical axis (SVA) ≥50 mm over 6 years.
Methods:
Fatty infiltration ratio (FIR) of the erector spinae and multifidus was quantified on MRI. Multivariable logistic regression models were constructed, adjusting for age, sex, body mass index (BMI), and baseline C7 SVA. FIR was expressed per 5% increase to improve clinical interpretability.
Results:
Sagittal malalignment progression occurred in 56 participants (10.5%). In univariable analysis, baseline C7 SVA did not differ significantly between groups, whereas pelvic tilt was significantly higher in the progression group. In multivariable models, pelvic tilt independently predicted progression (OR 1.13 per 1° increase, 95% CI 1.08-1.18, p<0.001). Fatty infiltration of the erector spinae at T12/L1 was also an independent predictor (OR 2.01 per 5% increase, 95% CI 1.40-3.05, p<0.001), whereas multifidus fatty infiltration was not statistically significant after adjustment. Baseline C7 SVA was inversely associated with progression (OR 0.82 per 10 mm increase, 95% CI 0.81-1.00, p=0.0085). In model discrimination analysis, the AUC improved from 0.700 in the base model to 0.806 after adding pelvic tilt and erector spinae FIR.
Conclusions:
MRI-defined fatty infiltration of the erector spinae was independently associated with sagittal malalignment progression in the general population. Increased pelvic tilt and lower baseline SVA may reflect a compensated state prone to subsequent decompensation. These findings highlight the importance of early identification of latent sagittal imbalance and suggest that paraspinal muscle quality may serve as a target for preventive strategies in ASD.
Level Of Evidence:
Level II.
