Fasting left reproductive hormones largely alone

Fasting and calorie restriction are among the most used tools for changing body composition, and both have been linked to better insulin sensitivity and lipid profiles. A persistent worry is what eating less does to the endocrine system — whether cutting energy intake suppresses testosterone in men or disrupts the hormones governing the menstrual cycle in women. Anecdote and gym folklore fill that gap in both directions, with some claiming fasting raises androgens and others that it tanks them. This review set out to gather the randomised evidence and put numbers on the question.

The authors searched PubMed, Scopus and Google Scholar for parallel or crossover randomised controlled trials published up to April 2025. Eligible trials enrolled adults aged 18 or over and tested alternate-day fasting, time-restricted eating or full-day fasting, or calorie restriction of 20%-30% of energy intake. Outcomes were testosterone, free testosterone, SHBG, DHEA-S, FSH, LH, oestradiol, oestrogen, anti-Müllerian hormone, prolactin and progesterone, pooled with random-effects models. Certainty of evidence was rated using the GRADE approach. Fifteen randomised trials with 954 participants in total were included.

Both fasting and calorie restriction produced largely neutral hormone results, with no consistent evidence of harm or benefit. Fasting's effect on testosterone was not statistically significant (WMD −0.88 ng/mL, 95% CI −2.37 to 0.61, p = 0.24) and came with significant heterogeneity between trials. Calorie restriction likewise did not significantly change testosterone (WMD −0.02 ng/mL, 95% CI −0.13 to 0.08, p = 0.65), again with heterogeneity. Free testosterone was the single significant finding, falling by 4.9 pg/mL (95% CI −9.22 to −0.59, p = 0.02), though the authors note this rested on some of the smaller trials. No significant effects appeared for any of the other hormones measured.

This pushes back on the idea that eating in a window or running a moderate deficit reliably reshapes the reproductive endocrine system in either direction. It sits alongside evidence that these strategies do improve insulin sensitivity and lipids, suggesting metabolic and reproductive responses are not tightly coupled at these levels of restriction. The heterogeneity for testosterone and the small-trial basis of the free-testosterone result leave room for subgroup effects the pooled numbers hide. The authors are explicit that they did not measure symptoms, so a flat hormone panel does not rule out changes people might notice.

The reasonable reading is that moderate fasting or calorie restriction, as tested in these trials, appears endocrinologically unremarkable at the group level. That is useful because it defuses a common fear about dieting rather than offering a new benefit to chase. The one signal worth noting is a fall in free testosterone, and it is the least robust part of the analysis. Larger and longer trials in specific populations are needed before anyone can say who, if anyone, responds differently.