Multiexponential Analysis of the Water T2-Relaxation in the Skeletal Muscle Provides Distinct Markers of Disease Activity Between Inflammatory and Dystrophic Myopathies.

Background: The monoexponential water T ₂ (T _2-mono ) is a proven biomarker of disease activity in neuromuscular disorders (NMDs). However, it lacks specificity, being elevated in the presence of several pathological processes and pathomorphological alterations in the muscle tissue.
Purpose: To investigate the multiexponential behavior of the water T ₂ -relaxation in the skeletal muscle of NMD patients, aiming to identify more sensitive and specific biomarkers of disease activity.
Study Type: Retrospective case-control.
Population: Thirty Duchenne muscular dystrophy and 114 inclusion body myositis patients and 55 control subjects.
Field Strength/sequence: 3T/Single-voxel proton spectroscopy ( ¹ H-MRS) and multispin-echo (MSE) imaging.
Assessment: Water T ₂ -decay curves generated from ¹ H-MRS data acquired at 14 echo-times were fitted to mono- and biexponential models and the adjusted R ² of each fit was computed. Additionally, T ₂ spectra were generated from a regularized inverse Laplace transform. For comparison, water T ₂ maps were generated from the MSE data. The performances of the different variables at identifying patients were assessed via receiver operating characteristic (ROC)-curve analysis.
Statistical Tests: Chi-square, Kruskal-Wallis, and Mann-Whitney with Bonferroni correction for multiple comparisons.
Results: T _2-mono was elevated in patients (P<0.05), but could not distinguish inclusion body myositis (IBM) from Duchenne muscular dystrophy (DMD). While 79% of IBM data presented a biexponential behavior, this was only 16% and 10% for DMD and control data, respectively (P<0.05). All T ₂ spectra presented an intermediate-T ₂ peak characterized by an elevated T ₂ in patients (P<0.05) and by a relative fraction that was abnormally smaller in IBM patients (P<0.05). Also, a long-T ₂ peak was exclusively observed in IBM patients. A combination of T ₂ -spectrum variables performed best at identifying patients.
Data Conclusion: T ₂ spectra not only provided more sensitive and specific markers of disease presence than the T _2-mono , but also allowed distinguishing IBM from DMD patients. This must reflect distinct predominant pathological alterations between these diseases, suggesting that these markers provide additional pathophysiological/histopathological information that are missing from T _2-mono .
Level of Evidence: 3 TECHNICAL EFFICACY STAGE: 3.
(© 2020 International Society for Magnetic Resonance in Medicine.)