What is the best reference site for a single MRI slice to assess whole-body skeletal muscle and adipose tissue volumes in healthy adults?

BACKGROUND Whole-body magnetic resonance imaging (MRI) is the gold standard for the assessment of skeletal muscle (SM) and adipose tissue volumes. It is unclear whether single-slice estimates can replace whole-body data. OBJECTIVE We evaluated the accuracy of the best single lumbar and midthigh MRI slice to assess whole-body SM, visceral adipose tissue (VAT), and subcutaneous adipose tissue (SAT). DESIGN Whole-body MRI was performed in 142 healthy adults aged 19-65 y [mean ± SD age: 37.0 ± 11.8 y; BMI (in kg/m(2)): 25.3 ± 5.9]. Single slices were taken at lumbar vertebrae L1-L5 plus intervertebral discs and the thigh (midthigh, 10 cm distally from the midthigh, and 10 cm proximally from the midthigh). The value of single-slice areas was also tested in a longitudinal study on 48 healthy volunteers during weight loss (8.2 ± 5.2 kg). RESULTS Cross-sectionally, all SM and adipose tissue single-slice areas correlated with total tissue volumes (P < 0.01). Because of the close associations between L3 areas and corresponding tissue volumes (r = 0.832-0.986, P < 0.01), this location was identified as the reference to estimate SM and adipose tissue in both sexes. SM, SAT, and VAT areas at L3 explained most of the variance of total tissue volumes (69-97%, with SEs of estimation of 1.96 and 2.03 L for SM, 0.23 and 0.61 L for VAT, and 4.44 and 2.47 L for SAT for men and women, respectively. There was no major effect on the explained variance compared with that for optimal slices. For SM, the optimal slice area was shown at midthigh. With weight-loss changes in total SM, VAT, and SAT, volumes were significantly different from those at baseline (SM changes: -2.8 ± 2.9 L; VAT changes: -0.7 ± 1.0 L; SAT changes: -5.1 ± 6.0 L). The area at L3 reflected changes in total VAT and SAT. To assess changes in total SM volumes, areas at midthigh showed the best evidence. CONCLUSION In both sexes, a single MRI scan at the level of L3 is the best compromise site to assess total tissue volumes of SM, VAT, and SAT. By contrast, L3 does not predict changes in tissue components. This trial was registered at clinicaltrials.gov as NCT01737034.