Correlation Between Quantitative MRI and Muscle Histopathology in Muscle Biopsies from Healthy Controls and Patients with IBM, FSHD and OPMD.

2020 
BACKGROUND Muscle MRI is increasingly used as a diagnostic and research tool in muscle disorders. However, the correlation between MRI abnormalities and histopathological severity is largely unknown. OBJECTIVE To investigate correlations between muscle MRI abnormalities and histopathological severity in healthy controls and patients with muscle disease. METHODS We performed quantitative MRI and histopathological analysis in 35 patients with inclusion body myositis, facioscapulohumeral muscular dystrophy or oculopharyngeal muscular dystrophy and 12 healthy controls. Participants contributed needle biopsies of the vastus lateralis and/or tibialis anterior, yielding 77 muscle biopsies with matched T1, T2 and TIRM MRI imaging. Muscle biopsies were evaluated with a semi-quantitative histopathology severity grading scale (range 0-12) and an inflammation severity grading scale (range 0-3). RESULTS In muscle disease, histopathology sum scores ranged from 0 to 11 and correlated significantly with fat percentage as measured on MRI (Spearman's rho = 0.594, p < 0.001). Muscle edema on muscle MRI was associated with increased amounts of inflammation (p < 0.001). Mild abnormalities occured in 95% of control biopsies and were more pronounced in tibialis anterior (median sum score of 1±1 in vastus lateralis and 2±1 in tibialis anterior (p = 0.048)). CONCLUSION In muscle disease, fatty infiltration on MRI correlates moderately with muscle histopathology. Histopathological abnormalities can occur prior to the onset of fatty infiltration. In middle-aged controls, almost all biopsies showed some histopathological abnormalities. The findings from this study may facilitate the choice for appropriate imaging sequences as outcome measures in therapeutic trials.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    22
    References
    3
    Citations
    NaN
    KQI
    []