Reducing energy availability in male endurance athletes: a randomized trial with a three-step energy reduction

BACKGROUND Low energy availability (EA) can be detrimental for athlete health. Currently, it is not known what the threshold for low EA in men is, and what effects it may have on performance. METHODS This study was set to determine potential effects of low EA by modulating male participants’ exercise energy expenditure and controlling energy intake and consequently manipulating EA in three progressive stages (reducing EA by 25%%, and 75 %). Performance was measured with three specific tests for explosive power, endurance, and agility. Blood was drawn, resting energy expenditure was monitored and two questionnaires were repetitively used to address any changes in eating behaviors and well-being. RESULTS Repeated measured design showed poorer performance (power output 391.82 ± 29.60 vs. 402.5 ± 40.03 W, p = 0.001; relative power output 5.53 ± 0.47 vs. 5.60 ± 0.47 W/kg, p = 0.018; explosive power 0.28 ± 0.04 vs. 0.32 ± 0.05 m, p = 0.0001, lactate concentration 7.59 ± 2.29 vs 10.80 ± 2.46 mmol/L, p = 0.001). the quartile range for testosterone was lower (2.33 ± 1.08 vs. 2.67 ± 0.78, p = 0.026) and there was a tendency for lower triiodothyronine (4.15 ± 0.61 vs. 4.46 ± 0.54 pmol/L, p = 0.072). Eating behaviors and well-being were worse (46.64 ± 7.55 vs. 24.58 ± 7.13, p = 0.011 and 15.18 ± 2.44 vs. 17.83 ± 3.54, p = 0.002). The intervention also resulted in lower body fat (8.44 ± 3.15 vs. 10.2 ± 2.5%, p = 0.013). CONCLUSIONS Analysis showed that most of the negative effects occurred in the range of 9–25 kCal·kg∙FFM·d−1. This is the range where we suggest a threshold for LEA in men could be. Reducing EA impaired explosive power first, then endurance. It was associated with a reduction in testosterone, triiodothyronine and there was a tendency for reduced IGF-1, but hormones were more resilient to changes in EA. Psychological assessment of eating behaviors and well-being proved to be very useful, whereas monitoring resting energy expenditure did not