Nutrition & DietResearch PaperPaywall

Time-Restricted Eating Slowed Forearm Bone Loss During Weight Loss in Postmenopausal Women

In a 24-week trial of 47 analyzed women, adding a sub-9-hour eating window to calorie restriction spared radius bone density but not hip bone, alongside gut microbiota shifts.

Friday, November 20, 2026 1 view
Published in Am J Clin Nutr
A middle-aged woman at a kitchen table eating a morning meal beside a wall clock showing an early hour

Summary

Losing weight often costs bone, a real concern for women after menopause. This 24-week randomized trial tested whether eating within a window of under 9 hours, on top of a 500-calorie daily cut, changes that. Forty-nine postmenopausal women with overweight or obesity were assigned to calorie restriction alone or to time-restricted eating plus calorie restriction. Both groups lost about 6.8 kg and lost similar hip bone density. At the forearm (radius), the time-restricted group lost less bone: density rose slightly versus a drop with calorie restriction alone. Their gut bacteria also shifted differently, and those shifts tracked with forearm bone changes. Blood bone turnover markers did not change. This is a small trial with a bone-density endpoint, so the microbiota link is an association, not proof of cause. The hip, a key fracture site, was not protected.

Detailed Summary

Weight loss improves metabolic health but typically costs bone, which matters for postmenopausal women already facing accelerated bone loss and fracture risk. Time-restricted eating (TRE) influences circadian rhythms and the gut microbiota, both of which have been linked to bone biology. Whether TRE changes bone loss during calorie restriction (CR) was unknown.

This randomized controlled trial enrolled 49 postmenopausal women aged 50-79 with a BMI of 25 or higher. For 24 weeks they followed either CR (-500 kcal/day) or TRE (eating window under 9 hours) plus CR. Bone mineral density (BMD) and bone quality were measured at weeks 0 and 24, stool samples at weeks 0, 12 and 24, and blood at weeks 0 and 24. Results covered 21 women in CR and 26 in TRE+CR.

The TRE+CR group cut its eating window by about 4 hours, with 94% adherence (88% in CR). Weight fell about 6.8 kg in both groups. Hip BMD declined similarly in both, with no between-group difference. At the radius, a non-weight-bearing site, TRE+CR attenuated loss: +0.009 g/cm2 versus -0.020 g/cm2 with CR alone. Gut microbiota composition shifted within TRE+CR and diverged from CR by week 12. Changes in microbial guild ratios correlated with radius BMD changes, and baseline guilds predicted radius preservation (AUC 0.73, wide confidence interval 0.54-0.92). Visceral fat fell more with TRE+CR. Serum bone turnover markers were unchanged, and glycemic indices improved equally in both groups.

The findings suggest TRE may offer a bone-sparing signal at some skeletal sites during weight loss, with the microbiota as a candidate mechanism. Because turnover markers did not move, how the effect arises remains unexplained.

Caveats are substantial: the trial is small, the hip was not protected, the microbiota findings are correlational, and the radius result came from a modest sample. Larger, longer studies with fracture-relevant endpoints are needed.

Key Findings

  • TRE+CR attenuated radius BMD loss versus CR alone: +0.009 vs -0.020 g/cm2 over 24 weeks in 47 analyzed women.
  • Hip BMD fell similarly in both groups (about -0.010 g/cm2); TRE did not protect the hip.
  • Gut microbiota diverged between groups by week 12, and guild-ratio changes correlated with radius BMD changes (association only).
  • Weight loss was similar (-6.8 kg), but visceral fat dropped more with TRE+CR; glycemic improvements did not differ.
  • Serum bone turnover markers were unchanged, leaving the mechanism of the radius effect unexplained.

Methodology

Randomized controlled trial of 49 postmenopausal women (BMI 25 or higher, ages 50-79) assigned to 24 weeks of CR (-500 kcal/day) or TRE (under 9 hours) plus CR; 47 were analyzed. Outcomes were BMD, bone quality, fecal microbiota (weeks 0, 12, 24) and blood markers. Microbiota-bone links came from correlation and subgroup prediction analyses.

Study Limitations

This summary is based on the abstract only. The trial is small, lasted 24 weeks, used BMD rather than fractures, and showed no hip benefit; the radius effect and microbiota correlations (AUC confidence interval 0.54-0.92) are exploratory and do not establish causation. Unchanged turnover markers leave the mechanism unclear.

Enjoyed this summary?

Get the latest longevity research delivered to your inbox every week.

Enter your email to subscribe: