Keto, fasting, and energy balance: positions that changed with the data

Few areas of nutrition have reversed as publicly as low-carbohydrate and fasting strategies. Positions held with confidence a decade ago, that ketogenic diets drive fat loss through insulin suppression, that extended fasts are necessary for cellular cleanup, that metformin is a longevity drug for healthy people, have each been revised as controlled data arrived. The revisions are more instructive than the original claims.

What the feeding studies showed

The metabolic ward, where every calorie eaten and expended is measured, is where the insulin model of obesity has been put to a controlled test. In a National Institutes of Health study published in 2016, 17 men with overweight or obesity lived on a ward and ate a high-carbohydrate baseline diet for four weeks, then switched to a ketogenic diet with the same calories for another four weeks [1]. Energy expenditure measured in the respiratory chamber rose briefly after the switch, by roughly 100 kcal a day at first, and averaged about 57 kcal a day higher across the ketogenic phase. Body fat loss did not speed up; it slowed during the ketogenic weeks, and the men lost fat-free mass as protein use rose [1]. A later reanalysis by other investigators argued that a second method, doubly labeled water, showed a larger rise in expenditure outside the chamber; Hall's group replied that the calculation did not account for the energy imbalance each diet produced [2]. Either way, the measured difference was tens of calories, not the hundreds the insulin model predicted, and fat loss did not accelerate.

A second ward study from the same group, published in 2019, isolated the variable most diets change without saying so. Twenty adults alternated between an ultra-processed diet and an unprocessed diet for two weeks each, with the meals matched for presented calories, sugar, fat, fiber, and macronutrients, and the participants ate as much or as little as they wanted [3]. On the ultra-processed diet they ate about 508 kcal more a day and gained 0.9 kg; on the unprocessed diet they lost 0.9 kg [3]. The macronutrient ratio was held constant, and intake still moved by 500 calories a day, so food form carried the effect.

Fasting, sized correctly

Time-restricted eating has fared similarly under testing. In the TREAT trial, published in 2020, 141 adults with overweight or obesity were randomized to either a 16:8 eating window (noon to 8 pm) or three structured meals a day for 12 weeks; 116 provided data [4]. The time-restricted group lost 0.94 kg, but the difference from the control group, 0.26 kg, was not significant. In a subset of 50 participants who had body composition measured in person, appendicular lean mass index fell more in the time-restricted group (between-group difference 0.16 kg/m², P = .005), which the authors flagged as a concern needing follow-up [4].

Smaller, shorter trials of early time-restricted eating have shown metabolic effects in a narrow setting. In a five-week crossover trial, eight men with prediabetes ate all their food in a six-hour window ending before 3 pm, with calories set to hold weight steady; insulin sensitivity, blood pressure, and oxidative stress markers improved compared with a 12-hour window [5]. That is a supervised, weight-stable, eight-person study, and its result has not been shown at scale.

Human data on autophagy during fasting are thinner than the popular account suggests. In one study, volunteers fasted for up to four days; autophagy activation in their blood cells was detectable only in neutrophils, and measuring the change in flux required culturing the cells with a protease inhibitor [6]. Most of what is said about fasting and autophagy comes from mouse and cell work. None of this supports multi-day fasts as a default practice for older adults, for whom lean mass is the resource a fast spends first.

Metformin in people without diabetes

The metformin case ran the same course. In a 12-week trial published in 2019, 53 older adults (mean age about 62) were randomized to metformin or placebo during supervised aerobic exercise training [7]. Metformin blocked the rise in skeletal muscle mitochondrial respiration that exercise produced in the placebo group, and whole-body insulin sensitivity did not improve on average in the metformin group because some participants improved and others got worse [7]. The authors concluded that more work was needed before metformin could be recommended to slow aging in healthy people.

Interpretation

The current evidence supports a plainer account. Diets work when they reduce intake in a way a person can sustain, and ketogenic diets are one such tool. They have a documented role in type 2 diabetes: in a one-year, non-randomized study of 262 adults on a very low-carbohydrate program with remote medical supervision, HbA1c, weight, and diabetes medication use all fell, with no change in 87 usual-care controls; the study was funded by the company running the program and the groups were not randomized [8]. Fasting windows are a scheduling tool with modest measured effects on weight. In both cases, the effect sizes in controlled studies are small and the adherence effects are large, and the ward data set the ceiling on what a diet's mechanism can deliver.

Sources

1. Hall KD, Chen KY, Guo J, et al. Energy expenditure and body composition changes after an isocaloric ketogenic diet in overweight and obese men. American Journal of Clinical Nutrition, 2016;104(2):324-333. https://doi.org/10.3945/ajcn.116.133561
2. Hall KD, et al. Mystery or method? Evaluating claims of increased energy expenditure during a ketogenic diet. PLoS ONE, 2019. https://pmc.ncbi.nlm.nih.gov/articles/PMC6901214
3. Hall KD, et al. Ultra-processed diets cause excess calorie intake and weight gain: an inpatient randomized controlled trial of ad libitum food intake. Cell Metabolism, 2019. https://doi.org/10.1016/j.cmet.2019.05.008
4. Lowe DA, et al. Effects of time-restricted eating on weight loss and other metabolic parameters in women and men with overweight and obesity: the TREAT randomized clinical trial. JAMA Internal Medicine, 2020;180(11):1491-1499. https://doi.org/10.1001/jamainternmed.2020.4153
5. Sutton EF, et al. Early time-restricted feeding improves insulin sensitivity, blood pressure, and oxidative stress even without weight loss in men with prediabetes. Cell Metabolism, 2018. https://doi.org/10.1016/j.cmet.2018.04.010
6. Pietrocola F, et al. Metabolic effects of fasting on human and mouse blood in vivo. Autophagy, 2017;13(3):567-578. https://doi.org/10.1080/15548627.2016.1271513
7. Konopka AR, et al. Metformin inhibits mitochondrial adaptations to aerobic exercise training in older adults. Aging Cell, 2019;18(1):e12880. https://doi.org/10.1111/acel.12880
8. Hallberg SJ, et al. Effectiveness and safety of a novel care model for the management of type 2 diabetes at 1 year: an open-label, non-randomized, controlled study. Diabetes Therapy, 2018. https://doi.org/10.1007/s13300-018-0373-9