Multiparametric Aging Study Across Adulthood in the Leg Through Quantitative MR Imaging, ¹H Spectroscopy, and ³¹P Spectroscopy at 3T

Abstract

BackgroundImproved characterization of healthy muscle aging is needed to establish early biomarkers in age‐related diseases.PurposeTo quantify age‐related changes on multiple MRI and clinical variables evaluated in the same cohort and identify correlations among them.Study TypeProspective.Population70 healthy subjects (30 men) from 20 to 81 years old.Field Strength/Sequence3T/water T2 (multiecho SE, multi‐TE STEAM), water T1 (GRE MR Fingerprinting), fat‐fraction (multiecho GRE, multi‐TE STEAM), carnosine (PRESS), multicomponent water T2 (ISIS‐CPMG SE train), and 31P pulse‐acquire spectroscopy.AssessmentAge‐ and sex‐related changes on: Imaging: fat‐fraction (FFMRI), water T1 (T1‐H2O), and T2 (T2‐H2O‐MRI) and their heterogeneities ΔT1‐H2O and ΔT2‐H2O‐MRI in the posterior compartment (PC) and anterior compartment (AC) of the leg. 1H spectroscopy: Carnosine concentration, pH, water T2 components (T2‐H2O‐CPMG), fat‐fraction (FFMRS), and water T2 (T2‐H2O‐MRS) in the gastrocnemius medialis. 31P spectroscopy: Phosphodiesters (PDE), phosphomonoesters, inorganic phosphates (Pi), and phosphocreatine (PCr) normalized to adenosine triphosphate (ATP) and pH in the calf. Clinical evaluation: Body‐mass index (BMI), gait speed (GS), plantar flexion strength, handgrip strength (HS), HS normalized to wrist circumference (HSnorm), physical activity assessment. Statistical TestsMultilinear regressions with sex and age as fixed factors. Spearman correlations calculated between variables. Benjamini–Hochberg procedure for false positives reduction (5% rate). A P < 0.05 significance level was used.ResultsSignificant age‐related increases were found for BMI (ρAge = 0.04), HSnorm (ρAge = −0.01), PDE/ATP (ρAge = 2.8 × 10−3), Pi/ATP (ρAge = 2.0 × 10−3), Pi/PCr (ρAge = 0.3 × 10−3), T2‐H2O‐MRS (ρAge = 0.051 msec), FFMRS (ρAge = 0.036) the intermediate T2‐H2O‐CPMG component time (ρAge = 0.112 msec), and fraction (ρAge = −0.3 × 10−3); and in both compartments for FFMRI (ρAge = 0.06, PC; ρAge = 0.06, AC), T2‐H2O‐MRI (ρAge = 0.05, PC; ρAge = 0.05, AC; msec), ΔT2‐H2O‐MRI (ρAge = 0.02, PC; ρAge = 0.02, AC; msec), T1‐H2O (ρAge = 1.08, PC; ρAge = 1.06, AC; msec), and ΔT1‐H2O (ρAge = 0.22, PC; ρAge = 0.37, AC; msec). The best age predictors, accounting for sex‐related differences, were HSnorm (R2 = 0.52) and PDE/ATP (R2 = 0.44). In both leg compartments, the imaging measures and HSnorm were intercorrelated. In PC, T2‐H2O‐MRS and FFMRS also showed numerous correlations to the imaging measures. PDE/ATP correlated to T1‐H2O, T2‐H2O‐MRI, ΔT2‐H2O‐MRI, FFMRI, FFMRS, the intermediate T2‐H2O‐CPMG, BMI, Pi/PCr, and HSnorm.Data ConclusionOur multiparametric MRI approach provided an integrative view of age‐related changes in the leg and revealed multiple correlations between these parameters and the normalized HS.Level of Evidence1Technical EfficacyStage 3