Abstract
OBJECTIVE: Weight/fat mass gain to support resumption of normal menses can be an important treatment goal for female athletes with amenorrhoea (AMEN) or oligomenorrhoea (OLIGO) related to relative energy deficiency in sport (REDs). However, targets based on body mass index (BMI) or per cent expected body weight (%EBW) may be less applicable in athletes, who typically have greater lean mass compared with the general population. Our aim was to determine if dual-energy X-ray absorptiometry (DXA)-derived %fat Z-score can detect menstrual status in female athletes and if a %fat Z-score cut-off of <-1.0 is a better indicator than standard BMI (in those >20 years) or %EBW (in those ≤20 years) thresholds. METHODS: 388 female athletes (age 15-30 years, Tier ≥2) with DXA scans were classified as AMEN (n=159), OLIGO (n=84) or naturally menstruating (NM, n=145) using clinical records. We compared %fat Z-score and BMI or %EBW values across menstrual groups and calculated sensitivity and specificity for predicting AMEN and the combined OLIGO and AMEN groups (OLIGO/AMEN) using traditional BMI or %EBW risk cut-offs versus a risk cut-off of <-1.0 for %fat Z-score. RESULTS: %fat Z-score was superior to traditional BMI or %EBW thresholds in discriminating between menstrual status groups. Using a %fat Z-score cut-off <-1.0 improved sensitivity (p<0.0001) for predicting AMEN (68.9%) and OLIGO/AMEN (57.7%) compared with the sensitivity of traditional BMI or %EBW cut-offs for AMEN (29.3%) and OLIGO/AMEN (25.9%). However, %fat Z-score <-1.0 did not improve specificity for predicting AMEN (75.4%) nor OLIGO/AMEN (79.9%) compared with the specificity of traditional BMI or %EBW cut-offs for AMEN (83.9%; p=0.0078) and OLIGO/AMEN (85.2%; p=0.14). CONCLUSIONS: When attempting to resume normal menstruation, achieving a %fat Z-score ≥-1.0, rather than using BMI or %EBW targets, may be a better goal for athletes with REDs-related amenorrhoea or oligomenorrhoea.