High-Speed T2 -Corrected Multiecho Magnetic Resonance Spectroscopy for Quantitatively Detecting Skeletal Muscle Fatty Infiltration and Predicting the Loss of Ambulation in Patients With Duchenne Muscular Dystrophy.

ABSTRACT

BACKGROUND: High-speed T2 -corrected multiecho MRS (HISTO-MRS) is emerging as a quantitative modality for detecting muscle fat infiltration (MFF). However, the predictive value of HISTO-MRS for the loss of ambulation (LoA) in Duchenne muscular dystrophy (DMD) is unknown. PURPOSE: To determine the feasibility of HISTO-MRS for assessing MFF in DMD and further identify the predictive value of HISTO-MRS for the LoA. STUDY TYPE Prospective. SUBJECTS: A total of 134 DMD boys (9.20 ± 2.43 years old) and 21 healthy boys (9.25 ± 2.10 years old). FIELD STRENGTH/SEQUENCE A 3 T, fast spin echo T1 -weighted imaging (T1 WI), two-point-Dixon gradient echo sequence (2-pt-Dixon) and HISTO-MRS. ASSESSMENT Subjective T1 WI fat grades by three radiologists, ROI analysis for MFF on 2 pt-Dixon (Dixon MFF) and MFF on HISTO-MRS (HISTO MFF) by two radiologists. Clinical motor function North Star Ambulatory Assessment, 10-m run/walk time, Gowers maneuver, and time to four-stairs climb and descend. STATISTICAL TESTS Spearman rank correlation was used to assess the relation of fat filtration assessments and motor ability. Bland-Altman plots was performed to determine the agreement of HISTO MFF and Dixon MFF. Receiver operating characteristic (ROC) curves were analyzed to determine the discriminating ability of above MRI modalities for ambulatory and nonambulatory DMD. Logistic regression was used to identify the predictor of LoA. Variables with P < 0.05 in univariate logistic regression analysis were entered into the multivariate logistic regression model. RESULTS: HISTO MFF was significantly correlated with Dixon MFF. Bland-Altman plots show good agreement of HISTO MFF and Dixon MFF. ROC curves indicated that HISTO MFF show similar discrimination of LoA for DMD with Dixon MFF but better value than T1WI fat grades. Logistic regression showed that HISTO MFF was an independent predictor for LoA. DATA CONCLUSION: HISTO-MRS is a potential quantitative method for assessing fat infiltration and shows predictive value for LoA in DMD patients. TECHNICAL EFFICACY Stage 5.