Effects of Exercise on Bone Marrow Adipose Tissue in Children With Overweight/Obesity: Role of Liver Fat

Abstract

Context: Exercise reduces adiposity, but its influence on bone marrow fat fraction (BMFF) is unknown; nor is it known whether a reduction in liver fat content mediates this reduction.

Objectives: This work aimed to determine whether incorporating exercise into a lifestyle program reduces the lumbar spine (LS) BMFF and to investigate whether changes in liver fat mediate any such effect.

Methods: Ancillary analysis of a 2-arm, parallel, nonrandomized clinical trial was conducted at primary care centers in Vitoria-Gasteiz, Spain. A total of 116 children with overweight/obesity were assigned to a 22-week family-based lifestyle program (control group [n = 57]) or the same program plus an exercise intervention (exercise group [n = 59]). The compared interventions consisted of a family-based lifestyle program (two 90-minute sessions/month) and the same program plus supervised exercise (three 90-minute sessions/week). The primary outcome examined was the change in LS-BMFF between baseline and 22 weeks, as estimated by magnetic resonance imaging. The effect of changes in hepatic fat on LS-BMFF were also recorded.

Results: Mean weight loss difference between groups was 1.4 ± 0.5 kg in favor of the exercise group. Only the children in the exercise group experienced a reduction in LS-BMFF (effect size [Cohen d] -0.42; CI, -0.86 to -0.01). Importantly, 40.9% of the reductions in LS-BMFF were mediated by changes in percentage hepatic fat (indirect effect: β=-0.104; 95% CI, -0.213 to -0.019). The effect of changes in hepatic fat on LS-BMFF was independent of weight loss.

Conclusion: The addition of exercise to a family-based lifestyle program designed to reduce cardiometabolic risk improves bone health by reducing LS-BMFF in children with overweight or obesity. This beneficial effect on bone marrow appears to be mediated by reductions in liver fat.

Keywords: MAFLD; MASLD; NAFLD; bone health; fatty liver; lifestyle; obesity; youth.

Figure 1.

Flowchart of children participating in the study. Control group: children participating in the family-based lifestyle and psychoeducation program; exercise group: children participating in the latter plus exercise training. BMFF, bone marrow fat fraction.

Figure 2.

Lumbar spine bone marrow fat fraction before (Pre) and after (22 weeks) participation in the family-based lifestyle and psychoeducation program (control, black), or the same plus exercise training (exercise group, red); results are for every participating child. The ends of the boxes in the box plots are located at the first and third quartiles; the black line in the middle denotes the median. Whiskers extend to the upper and lower adjacent values, the location of the furthest point within a distance of 1.5 interquartile ranges from the first and third quartiles. The parallel line plot shows one vertical line for each participant, extending from their baseline to their 22-week value. Descending lines indicate a reduction in the lumbar spine bone marrow fat fraction, whereas ascending lines indicate an increase. Pretest values are placed in ascending order for the control group and descending order for the exercise group. Intragroup (pre vs post) differences were calculated using the paired t test. Intergroup (control vs exercise) differences (post minus pre) were calculated by one-way analysis of covariance adjusted for baseline value, age, and sex. The data used for the figure can be found in eTable 1.

Figure 3.

A, Mediation analysis. Effects of the intervention on changes in bone marrow fat fraction (BMFF) through changes in percentage hepatic fat, adjusting for confounders (baseline values, age, and sex). B, Lumbar spine BMFF before (Pre) and after (22 weeks) for every child with overweight/obesity who experienced a reduction (d-Cohen ≥ 0.2) (responders), or not (nonresponders, d-Cohen < 0.2), in percentage hepatic fat by the end of the intervention. The ends of the boxes in the box plots are located at the first and third quartiles; the black line in the middle denotes the median. Whiskers extend to the upper and lower adjacent values, the location of the furthest point within a distance of 1.5 interquartile ranges from the first and third quartiles. The parallel line plot contains one vertical line for each participant, extending from their baseline to their 22-week value. Descending lines indicate a reduction in the lumbar spine bone marrow fat fraction whereas ascending lines indicate an increase. Pretest values are placed in ascending order for non-responders and descending order for responders. Intragroup (pre vs post) differences were calculated using the paired t test. Intergroup (control vs exercise) differences were calculated by one-way analysis of covariance adjusted for baseline value, age, and sex. The data used for Fig. 3B can be found in eTableS2.