To assess the relationship between different aspects of water T2 relaxation time and age in leg muscles in Duchenne Muscular Dystrophy (DMD).Quantitative MRI is gaining momentum as a possible outcome measure in clinical trials for neuromuscular diseases.¹ Two major hallmarks of muscle damage are inflammation and replacement of muscle tissue by fat, both of which can be assessed by MRI. While fat infiltration is known to increase with age,² the relation between T2 and age is less equivocal. The interpretation of muscle T2 in muscular dystrophies is complicated by the disease progression, as both inflammation and increased fat content result in a longer T2. Different approaches of extracting the underlying disease activity have been proposed, including MR spectroscopy and water-fat separation.^3,4 Here we measure T2 in two muscles of the lower leg with different involvement, the soleus (SOL) and tibialis anterior (TA), using a tri-exponential fitting of the T2 decay, and analyze its relation with age.MR imaging of the right lower leg was performed in 21 boys with DMD (mean age 10.1±2.8 years, range 5-16 years) and 12 age- and sex-matched controls (10.3±2.7 years, range: 5-14 years) using a 32-element coil at 3T (Ingenia, Philips, Best, The Netherlands). Axial TSE images (17 echoes; TR/TE/ΔTE 3000/8/8ms; voxel size 1.4×1.8×10 mm³; no fat suppression, 5 slices) and a B_1⁺ map (AFI, TR1/TR2/TE 30/150/2.9ms, voxel size 2.8×2.8x24 mm³) were acquired. In 5 additional DMD patients (10.3±1.8 years, range 8-13 years), single voxel MRS of the SOL was added to the protocol (STEAM; 16 echoes; TR/TE/ΔTE 5000/12/15ms; 4 averages; voxel size 20×20×40 mm³) with offline fitting of the water peak (jMRUI software, v5.2). The muscle T2 was calculated in each voxel with a tri-exponential fitting using Matlab (Mathworks, USA), with two components for the fat T2 and one for muscle.⁴ Manual segmentation of the SOL and TA muscles was performed on a single slice of the image with the shortest echo time and served as a mask for the T2 map (Fig. 1). During this step, the region of interest was evaluated as unsuitable for further quantitative analysis due to extensive pulsation- or motion-artefacts in 8 subjects for the TA (6 patients, 2 controls) and in 4 patients for the SOL. In the next step, voxels with B_1⁺ outside the range of 85-130% of the target value were omitted.⁴ Average T2 values as well as T2-heterogeneity (defined as the coefficient of variation) per muscle were calculated. Statistical analysis was performed with SPSS (v20, IBM). The significance level was set to 0.05.For the SOL, 22% (range 0-75%) of the voxels were outside the accepted B_1⁺ range while for the TA all voxels were included except for 2 patients, for whom the whole TA was excluded. One patient dataset was excluded due to a high fat fraction (75%), which made the tri-exponential fit unreliable. Although the mean water T2 was slightly higher in patients compared to controls, this did not reach statistical significance (independent t-test, SOL: 37.02 ± 4.0ms vs. 35.27 ± 1.23ms, TA: 36.97 ± 2.15ms vs. 35.40ms ± 0.88ms). The heterogeneity in T2 values was significantly higher for patients in the SOL (7.5±4.3% vs. 4.3±1.1%, p 0.014) but not in the TA (5.3±1.6% vs. 4.1±1.4%, p 0.1). There was a positive correlation with age for T2-heterogeneity (r^2 = 0.609, p<0.001) and no correlation for mean T2 in DMD patients (Fig. 2). The T2 of the SOL measured using MRS showed a correlation trend with the water-T2 value derived from imaging (r^2 = 0.806, p 0.06).Using the tri-exponential fitting method, we found slightly, but not significantly elevated T2 values in DMD patients, which has been described before and may reflect the chronic inflammation.² We showed that the muscle T2 values measured with MRS and MRI tend to correlate in patients. The two populations differed in the heterogeneity of the T2 values in the SOL, while the lack of heterogeneity for the TA might be attributed to the high voxel exclusion rate. The wide span of T2 values within small regions of interest can be explained by the coexistence of ongoing disease activity (higher T2 values) and advanced fibrosis (lower T2 values). Our findings are in accordance with previous research in the dog model and in the upper extremity of DMD patients.^5,6 The relationship between muscle T2 and age is complex and interpretation of T2 alone can be insufficient. T2-heterogeneity should also be taken into consideration when using the water T2 of the diseased muscle as an outcome measure for therapeutic interventions.