0738

Multi-parametric ageing study on 51 subjects in the lower leg by ¹H water T₁ MR fingerprinting, multi-compartment water T₂, fat fraction and ³¹P MRS

Alfredo Liubomir Lopez Kolkovsky¹, Béatrice Matot¹, Harmen Reyngoudt¹, Benjamin Marty¹, Ericky Caldas de Almeida Araujo¹, and Yves Fromes¹
¹NMR Laboratory , Neuromuscular Investigation Center, Institute of Myology, Paris Cedex 13, France

Synopsis

Keywords: Muscle, Aging, Multi-contrast, Relaxometry, MR Fingerprinting

Aging is a multi-factorial process and studies in the lower leg are scarce. We performed a multi-contrast protocol in 51 volunteers from 20 to 81 y.o. We found an age-related increase of muscle water T₁(water-fat separation,MR fingerprinting), fat fraction, water T₂ and T₂ heterogeneity in the anterior and posterior compartments. Phosphodiesters and mitochondrial stress biomarkers also increased with age. Through bi-compartment water T₂ CPMG measures, age-related increases of the long water T₂ relative fraction were observed relative to the short water T₂ (T_{2-H2O-CPMG-short}) fraction, but T_{2-H2O-CPMG-short} values were unaltered with age, suggesting inflammation with preservation of the intracellular water compartment.

Introduction

The progressive decline in muscle mass, strength and performance during ageing negatively affects the quality of life in the elderly and increases the risk of falls, disability and frailty. This decline is a multifactorial process whose mechanisms are incompletely understood. NMR allows evaluating anatomical, structural and physiological aspects of muscle tissue which could provide relevant biomarkers to evaluate muscle ageing non-invasively and early biomarkers for age-related muscle degeneration. Contrary to the quadriceps, ageing studies on muscle water T₂ (T_2-H2O-MRI), fat fraction (FF), intracellular acidity and phosphodiester (PDE) relative are scarce in the lower leg^1-3. Multi-compartment T₂ and water T₁ (T_1-H2O) measurements have demonstrated their interest in neuromuscular diseases^4,5 but have not yet been evaluated in the context of ageing. Here, we evaluated the effect of aging on muscle T_1-H2O, T_2-H2O-MRI and T₂ heterogeneity (ΔT_2-H2O-MRI), FF and cross-sectional area (CSA) in the lower leg. Multi-compartment water T₂ CPMG (T_2-H2O-CPMG) measures in the gastrocnemius and ³¹P MRS in the calf were also performed.

Methods

Experimental Setup
Experiments were done at 3T (Siemens Prisma) on 51 healthy subjects (20 to 81 y.o.). The RF coil used (RAPID Biomedical) combined a ¹H birdcage transmitter, a ¹H ¹⁸Rx phased-array receiver and a ³¹P ¹Tx/³Rx semi-cylindrical transceiver.
NMR measures
MRI
FF was measured using a 3-point Dixon method (3D GRE, TEs=2.75/3.95/7.55 ms, TR=10 ms, FA=3°, 48 slices, 1x1x5 mm³ resolution). T₂-weighted images were measured using a multi-spin-echo sequence (17 echoes, TE₁/ΔTE=9.5 ms, TR=3 s, 5 slices, 10 mm thickness, 1.3x1.3 mm² in-plane resolution) and T_2-H2O-MRI maps were estimated with a tri-exponential fitting procedure6. A MRF T1-FF sequence was employed (radial encoding, 1400 spokes train, 5 slices, 10 mm thickness, 1.4x1.4 mm² in-plane resolution) to generate T_1-H2O maps using a bi-component model⁷. Manual segmentation was done for individual muscles on 3 slices of the multi-spin-echo data.
Multi-compartment T₂
An ISIS-CPMG sequence with fat suppression was used to measure the water T₂ decay⁸ (150 echoes, 2.5 ms inter-echo spacing, 35x35x35 mm³ voxel size, TR=9 s). T_2-H2O-CPMG times and their relative fractions were estimated using a non-negative least squares algorithm8 (480 ms upper limit). A short (<50 ms) and a long (>50 ms) T_2-H2O-CPMG component are reported.
³¹P MRS
A pulse-acquire sequence was used (3 kHz bandwidth, 2048 complex points, TR=4 s, 64 averages). Spectral analysis was done using jMRUI (v5.2), AMARES and TopSpin (Bruker Medical) as described elsewhere⁹. Phosphocreatine (PCr), inorganic phosphate (Pi), PDE, phosphomonoesters (PME) and the gamma adenosine triphosphate (ATP) resonances were quantified. pH was estimated from the chemical shift between Pi and PCr¹⁰.
Statistical analysis
A linear model and ANOVA with gender and age as fixed factors were used for statistical analysis (R Studio), with p<0.05 considered significant.

Results

Figure 1 shows the participants’ demographics and clinical evaluation results. Figure 2 summarizes the changes related to age and gender in the MRI data. Overall, consistent age-related increases of FF, T_1-H2O, T_2-H2O-MRI and ΔT_2-H2O-MRI were observed in individual muscles, as well as sex-related changes, notably for T_2-H2O-MRI and CSA. Figure 3 shows the increase of these variables in the posterior and anterior compartments for each gender. As shown in figure 4, PDE/ATP and Pi/PCr increased with age but no changes were observed for PME/ATP and only sex-related differences were observed for pH. Interestingly, the relative fraction of long T_2-H2O-CPMG increased with age (figure 5A) but the short and long T_2-H2O-CPMG values did not (figure 5B,C). By ignoring the volunteers whose long T_2-H2O-CPMG reached the upper limit of the fit, a significant age-related decrease of long T_2-H2O-CPMG values was found (figure 5D).

Discussion

The age-related increase of FF and T_2-H2O-MRI in the lower leg agree with studies in the quadriceps^1,11. An increase of T_1-H2O-MRI with age was also found, similarly to another study where multiple T_1-H2O-MRI measurements strategies were employed². Unlike T_2-H2O-MRI, T_1-H2O-MRI was insensitive to gender but systematically presented lower R² values. The increase of T_2-H2O-MRI heterogeneity, the stability of the short T_2-H2O-CPMG component during ageing and the increase of the long T_2-H2O-CPMG relative fraction, could indicate an increase of the vascular compartments⁸, possibly due to chronic inflammation processes, fiber type II atrophy and fluid retention. The high variability of our long T_2-H2O-CPMG values preclude us to conclude on the impact of ageing (figure 5C-D) and requires further investigation. The age-related increase in PDE/ATP and Pi/PCr (and absence of PME/ATP change) are in agreement with previous studies in the thigh¹² and calf³, and suggest an increased membrane turn-over, possibly related to a degeneration or change of fiber topology, as well as increased mitochondrial stress.

Conclusion

Age-related structural and metabolic changes in the lower leg were observed, demonstrating the interest of a multi-parametric approach. Future studies could focus on the etiology of sarcopenia and, in animal studies, confront the NMR findings to histological data.

Acknowledgements

No acknowledgement found.

References

1. Codari M et al., JMRI. 2020; 51 :1117-1127

2. White JC et al., Diagnostics (Basel). 2022; 12(3) :584.

3. Krumpolec P et al., Front Physiol. 2020; 11 :644.

4. Araujo ECA et al., JMRI. 2021; 53(1) :181-189.

5. Marty B et al., Radiology. 2021; 300(3) :652-660

6. Azzabou N et al., JMRI. 2015; 41(3) :645-53.

7. Marty B & Carlier PC, MRM. 2019; 00 :1-14

8. Araujo ECA et al., Biophys J. 2014; 106 :2267–2274

9. Reyngoudt H et al., NMR Biomed. 2018; 31 :e3839.

10. Moon RB & Richards JH. J Biol Chem. 1973; 248 :7276-7278.

11. Azzabou N et al., Exp Gero. 2015; 70 :54-60

12. Reyngoudt H et al., 8th International Conference of the SCWD. Abstract 5-06. JCSM. 2015; 6(4): 398–509.

Figures

Figure 1. Participant demographics and density distribution by age and sex.

Figure 2. Correlations of T_1-H2O, T_2-H2O-MRI, ΔT_2-H2O-MRI, FF and CSA with age and sex. The posterior compartment included the soleus and gastrocnemius (gas.) medialis and lateralis. The anterior compartment included the tibialis anterior (tib. ant.) and posterior (tib. post.), extensor digitorum (ext. dig.). The compartments’ linear regression results are shown in figure 3. All age-related changes showed an increase with age. Statistical significance levels: *P <0.05, **P <0.005, ***P <0.0005.

Figure 3. Linear correlations with age and sex for the shown variables in the posterior (left) and anterior (right) compartments. The posterior compartment included the soleus and gastrocnemius. The anterior compartment included the tibialis anterior and posterior, extensor digitorum and fibular muscle. Significance for age and sex are shown for each variable. CSA: muscle cross-sectional area, FF: fat fraction. Straight lines denote the linear regression, shaded areas display the confidence bands. Statistical significance levels: *P <0.05, **P <0.005, ***P <0.0005.

Figure 4. Linear correlations with age and sex for PDE/ATP, PME/ATP, Pi/PCr and pH measured in the calf. Significance for age and sex are shown for each variable. PDE: phosphodiesters, PME: phosphomonoesters, ATP: Adenosine triphosphate, PCr: phosphocreatine, Pi: inorganic phosphate. Straight lines denote the linear regression, shaded areas display the confidence bands. Statistical significance levels: **P <0.005, ***P <0.0005.

Figure 5. Linear correlations with age and sex for the (A) relative fraction of the long T_2-H2O-CPMG component and the fitted (B) short and long (C) T_2-H2O-CPMG values, measured in the gastrocnemius med. In (D), the linear correlation on the long T_2-H2O-CPMG values was repeated as in (C) but the values reaching the upper limit of the fit (480 ms) were excluded and gender differences were not considered. Significance for age and sex are shown. Straight lines denote the linear regression, shaded areas display the confidence bands. Statistical significance levels: *P <0.05, ***P <0.0005.

Proc. Intl. Soc. Mag. Reson. Med. 31 (2023)

0738

DOI: https://doi.org/10.58530/2023/0738