ABSTRACT
Objective: To investigate differences in diffusion tensor imaging (DTI) parameters and proton density fat fraction (PDFF) in the spinal muscles of younger and older adult males. Methods: Twelve younger (19-30 years) and 12 older (61-81years) healthy, physically active male participants underwent T1W, T2W, Dixon and DTI of the lumbar spine. The eigenvalues (λ1, λ2, and λ3), fractional anisotropy (FA), and mean diffusivity (MD) from the DTI together with the PDFF were determined in the multifidus, medial and lateral erector spinae (ESmed, ESlat), and quadratus lumborum (QL) muscles. A two-way ANOVA was used to investigate differences with age and muscle and t-tests for differences in individual muscles with age. Results: The ANOVA gave significant differences with age for all DTI parameters and the PDFF (P < .01) and with muscle (P < .01) for all DTI parameters except for λ1 and for the PDFF. The mean of the eigenvalues and MD were lower and the FA higher in the older age group with differences reaching statistical significance for all DTI measures for ESlat and QL (P < .01) but only in ESmed for λ3 and MD (P < .05). Conclusions: Differences in DTI parameters of muscle with age result from changes in both in the intra- and extra-cellular space and cannot be uniquely explained in terms of fibre length and diameter. Advances in knowledge: Previous studies looking at age have used small groups with uneven age spacing. Our study uses two well defined and separated age groups.
ABSTRACT
OBJECTIVE: To validate MRI fat measurement protocols using purpose built test objects and by comparison with air-displacement plethysmography (ADP) whole-body fat measurements in non-obese subjects. METHODS: Test objects of known fat concentration were used to quantify the accuracy of the MRI measurements. 10 participants with a body mass index in the range 18-30 underwent whole-body MRI using two different Dixon-based sequences (LAVA Flex and IDEAL IQ) to obtain an estimate of their whole-body fat mass. The MRI determined fat mass was compared to the fat mass determined by ADP. RESULTS: MRI test object measurements showed a high correlation to expected fat percentage (r > 0.98). The participant MRI and ADP results were highly correlated (r = 0.99) but on average (mean ± standard deviation) MRI determined a higher fat mass than ADP (3.8 ± 3.1 kg for LAVA Flex and 1.9 ± 3.2 kg for IDEAL IQ). There was no trend in the difference between MRI and ADP with total fat mass. CONCLUSION: The good agreement between MRI and ADP shows that Dixon-based MRI can be used effectively as a tool in physiological research for non-obese adults. ADVANCES IN KNOWLEDGE: This work found that for ten non-obese subjects body mass index had no effect on the MRI determination of whole-body fat mass.
