Single-slice quantitative magnetic resonance imaging (MRI) of the thigh for differential assessment of fasting-related body composition changes.

AIM: The aim of this study was to assess the ability of single-slice quantitative thigh magnetic resonance imaging (MRI) to estimate whole-body (wb) composition and to detect fasting-related changes, providing potential surrogate markers. MATERIALS AND METHODS: This prospective substudy is based on a previously published randomised clinical trial with 66 non-obese participants in three matched intervention groups: control (n=20), alternate-day fasting (n=23), and time-restricted feeding (n=23). Body composition (BC) analysis was performed before and after a month-long intervention with wbMRI, multi-echo Dixon thigh MRI and bioimpedance analysis (BIA). Thigh compartments were manually segmented on a single slice of the fat fraction (FF) map, including muscle area (MA) and intermuscular adipose tissue (IMAT) of extensor and flexor compartments, respectively. Relationships between thigh MRI, wbMRI, and BIA were modelled with simple linear regressions and sandwich estimators. Effects of fasting regimens on thigh measurements compared to the control group were evaluated with analysis of covariance. RESULTS: Extensor more than flexor MA was a strong predictor of wb muscle mass in wbMRI and BIA (R2 = 0.77 and 0.72, respectively). The ability of thigh FF and IMAT to estimate visceral fat volume was weak to moderate (R2 = 0.15 to 0.43). Fasting regimens did not demonstrate statistically significant changes of MA, FF, or IMAT in the extensor or flexor thigh compartment, compared to the control group (all P≥0.45). CONCLUSION: Our results suggest that single-slice thigh MRI can generally predict wb muscle mass and also visceral fat to a lesser degree. No fasting-related changes of tissue composition were detected, although certain trends were observed.