Williams · American journal of physiology. Endocrinology and metabolism 2015 · Randomized controlled trial · n=34

Magnitude of daily energy deficit predicts frequency but not severity of menstrual disturbances associated with exercise and caloric restriction.

Cited 152 times in the scientific literature.

Level 2 - randomized trial

Individual randomized controlled trial

PubMed 25352438 · doi:10.1152/ajpendo.00386.2013 · record verified 2026-08-29

What was done

Untrained, eumenorrheic women aged 18–30 years were evaluated over four menstrual cycles (one baseline and three intervention cycles) using controlled feeding and supervised exercise. Participants were randomized to an exercising control group (EXCON) or one of three exercising energy deficit groups: mild (ED1; -8 ± 2%), moderate (ED2; -22 ± 3%), or severe (ED3; -42 ± 3%). Measured outcomes included cycle length and urinary estrone-1-glucuronide, pregnanediol glucuronide, and midcycle luteinizing hormone.

What was found

Thirty-four subjects completed the study. Weight loss occurred in ED1 (-3.8 ± 0.2 kg), ED2 (-2.8 ± 0.6 kg), and ED3 (-2.6 ± 1.1 kg), compared to minimal change in EXCON (-0.9 ± 0.7 kg). The total sum of disturbances (luteal phase defects, anovulation, and oligomenorrhea) was greater in ED2 than EXCON, and greater in ED3 than both EXCON and ED1. Average percent energy deficit predicted disturbance frequency (f = 10.1, β = -0.48, r² = 0.23, P = 0.003) independent of weight loss. Deficit estimates associated with disturbances ranged from -22% (ED2; -470 kcal/day) to -42% (ED3; -810 kcal/day). Deficit magnitude did not predict disturbance severity.

Why it matters

This study experimentally demonstrates a dose-response relationship between the magnitude of caloric restriction during exercise and the frequency of menstrual cycle disruption in healthy women.

Limits

The sample size was small with 34 completers, limiting statistical power for specific disturbance subtypes. The intervention lasted only three cycles in untrained young women, so long-term outcomes, reversibility, and applicability to highly trained athletes were not evaluated.

Cited by