Randomised Controlled Trial of the Effects of Increased Energy Intake on Menstrual Recovery in Exercising Women With Menstrual Disturbances: the 'REFUEL' Study

The Female Athlete Triad is the condition characterized by the interrelationships among energy deficiency, menstrual dysfunction, and low bone density in exercising women. Increasing energy (ie, food) intake has shown some efficacy in case studies, retrospective cohort studies, and uncontrolled interventions in reversing oligomenorrhea. Protocol specifics including the amount of increased energy intake required for menstrual recovery, timeframe over which recovery occurs, and associated metabolic changes remain unclear. This randomized controlled trial aimed to investigate whether a 12-month intervention of increased energy intake leads to menstrual recovery in a cohort of women with severe exercise-associated menstrual disturbances. Women were eligible if in good health, aged 18 to 35 years, with body mass index between 16 and 25, with 2 hours per week of purposeful exercise, not currently dieting, not pregnant, not taking medication that could interfere with metabolic or reproductive hormones, with no menses in the prior 3 months or <6 cycles in the past 12 months. Participants were randomized into study groups, one of which increased energy intake for the duration of the 12 months, and a control group that maintained habitual exercise energy expenditure and energy intake. The nutritional intervention group gradually increased energy intake 20% to 40% above baseline energy requirements, which were determined using resting metabolic rate, daily energy expenditure, and thermic effect of food. Two primary outcomes were recorded, menstrual recovery defined by increased frequency of menses and improved menstrual function, and energy status as indicated by body weight and body composition. Body weight was measured biweekly, energy intake was self-reported monthly using diet logs, and body composition was analyzed using dual-energy x-ray absorptiometry. Occurrence ofmenses was self-reported throughout the study then corroborated by daily urinary reproductive hormones assessments in addition to blinded determinations of menstrual function by 2 experts. A total of 76 women were randomized, 40 to the nutritional intervention group and 36 to the control group, with 17 completing all 12 months of the study in the intervention group and 16 completing in the control group. The dropout rate was similar between both groups ( P = 1.0000; control 56%, intervention 58%). Increased energy intake had a positive effect on likelihood of experiencing menses ( P = 0.002) after controlling for baseline fat mass and menstrual status. Intent to treat analysis revealed that women in the intervention group were nearly twice as likely (hazards ratio, 1.91; 95% confidence interval, 1.27-2.89) to experience a menses during the study than those in the control group. Baseline fat mass significantly influenced likelihood of a participant experiencing a menses over the study period, with an increase of 1 kg at baseline increasing likelihood of menses by 8% (hazards ratio, 1.08; 95% confidence interval, 1.04-1.13). Subgroup analysis including only women with clear baseline menstrual status and adequate time in the study to evaluate change in menstrual function found that 64% of women in the intervention group had improved menstrual function compared with 19% in the control group (chi(2) = 11.2; P = 0.001). The intervention group experienced a significant increase in energy intake, body weight, percent body fat, and TT3 concentration ( P < 0.05) compared with the control group. The results of this study demonstrate menstrual recovery associated with increased energy intake of approximately 300350 kcal/d among exercising amenorrheic and oligomenorrheic women.