Evaluating lower limits of body fat percentage in athletes using DXA.
Level 4 - case-series / case-control
Cross-sectional observational study without an intervention or control comparison
PubMed 39869984 · doi:10.1016/j.jocd.2025.101564
What was done
In a cross-sectional study, 732 free-living participants with unrestricted food access (643 competitive athletes across 18 sports and 89 non-athletes; 454 males, mean age 20.9 ± 3.8 years; 278 females, mean age 21.3 ± 5.5 years) underwent whole-body DXA scans (Horizon A Hologic, software version 13.6.0.5) performed and analyzed by a single operator to establish the lowest DXA-estimated whole-body fat mass in athletes.
What was found
In males (n = 454), mean body fat percentage (BF%) was 18.2 ± 4.9% (range 10.3–37.2%), with basketball players lowest (15.3 ± 2.6%) and non-athletes highest (21.6 ± 5.4%) (1-way ANOVA F = 12.3; p < 0.0001); average absolute fat mass was 19.9 ± 8.0 kg (range 6.7–62.0 kg). In females (n = 278), mean BF% was 27.1 ± 5.1% (range 16.2–45.3%), with runners lowest (23.5 ± 3.5%) and non-athletes highest (31.7 ± 6.3%) (1-way ANOVA F = 12.9; p < 0.0001); average absolute fat mass was 18.9 ± 6.4 kg (range 9.2–49.7 kg). The observed lower limits of whole-body fat mass were approximately 10% for males and 16% for females.
Why it matters
Because DXA consistently produces higher body fat readings than other assessment modalities, these findings provide practical lower-boundary reference values for interpreting DXA scans in competitive athletes.
Limits
The cross-sectional design cannot determine whether these lower boundaries represent safe or physiologically healthy thresholds, nor can DXA differentiate essential fat embedded within lean soft tissue. Results may not generalize across different DXA hardware or software versions. Hydration and exact dietary intake were not controlled.