Comparative Evaluation of Imaging Techniques for Paraspinal Muscle Fat Quantification

Abstract

Introduction: Low back pain (LBP) is a prevalent health issue associated with morphological changes in paravertebral muscles. Magnetic resonance imaging (MRI) is effective in identifying muscle steatosis, with MR spectroscopy (MRS) considered the gold standard. However, MRS is limited by technical challenges, prompting interest in Dixon sequences for fat quantification. Materials and Methods: We conducted a study comparing Multi Echo VIBE Dixon with single-voxel MR spectroscopy in quantifying fat fractions in lumbar multifidus muscles of chronic LBP patients. Ninety-eight measurements from 49 patients were analysed. Qualitative and quantitative image analyses were performed, and statistical analysis was conducted using SPSS 27. Results: Significant positive correlation (correlation coefficient: 0.992, p < 0.001) was found between Dixon and MRS fat fraction measurements. Intra-class correlation coefficient was 0.991 (p < 0.001), indicating strong agreement. No proportion bias was observed. Factors such as age, sex, and spinal degeneration correlated positively with multifidus fat atrophy. Conclusion: T2*-corrected Dixon imaging aligns well with spectroscopic measurements, offering an accurate alternative for estimating paraspinal muscle fat content in chronic LBP patients. Further studies are needed to establish threshold values for Dixon imaging. Chronic LBP is associated with multifidus muscle atrophy and fat infiltration, highlighting the need for effective management strategies.

Keywords

• low back pain
• multifidus muscle
• mri
• dixon imaging
• mr spectroscopy