What a 10-Day Modified Fast Actually Does to Muscle

Extended fasting can cost you some body protein. The question is how much—and whether a falling "lean mass" number means your muscles are disappearing. In the only prospective 10-day trial designed to sort that out, most of the measured lean-soft-tissue loss appeared to be water, depleted glycogen, and emptied gut contents rather than protein-containing tissue. That is reassuring, but only within a narrow setting: 16 healthy men completed a medically supervised, calorie-containing Buchinger fast while exercising every day.

So clinicians who warn about muscle loss are pointing to a real problem, especially early in a fast. The leap from that warning to "a 10-day fast burns away your muscle" goes beyond the available data.

The trial behind the claim

Claire Laurens and colleagues' 2022 prospective trial followed 16 healthy adult men through a 10-day Buchinger modified fast. This was not a water-only fast. Participants consumed 200–250 kcal per day from honey, juice, and vegetable broth. They also did one hour of gymnastics and two hours of walking each day, and the protocol required enemas every other day.

The researchers used several indirect measures together: 24-hour urinary nitrogen excretion, extracellular-water estimates from bioelectrical impedance spectroscopy, and assumptions about glycogen depletion drawn from earlier biopsy literature. Their model estimated that lean soft tissue made up about 60% of total weight loss, while fat mass made up roughly 40%.

That first number sounds alarming until you ask what "lean soft tissue" contains. It includes skeletal muscle, but it also includes water, glycogen, organs, connective tissue, and material in the digestive tract. The investigators estimated that only about 42% of lost lean soft tissue was metabolically active, protein-containing tissue. The other 58% reflected water, glycogen depletion, and gut contents cleared by the protocol.

Glycogen explains part of the rapid early drop. The model assumed liver glycogen fell by about 90% and muscle glycogen by about 40%. Glycogen is stored with water, so draining it can move the scale—and a body-composition readout—quickly without representing equivalent loss of contractile muscle protein.

Protein breakdown appeared highest early

The study also tracked plasma 3-methylhistidine, a marker associated with breakdown of myofibrillar proteins. Its pattern fit the expected starvation physiology: an early rise followed by a decline around day 5, when ketosis had become established. The authors interpreted that decline as evidence of protein sparing, because ketones can reduce the body's need to make glucose from amino acids.

This is plausible, not conclusive. The trial measured plasma rather than urinary 3-methylhistidine, and the ELISA assay had an inter-assay coefficient of variation of about 10%. That is a meaningful limitation when the expected changes are small. A biomarker trend can support a physiological interpretation; it cannot prove exactly how much skeletal-muscle protein each participant lost.

The researchers did check whether their overall energy-balance arithmetic hung together. Accelerometry measured total energy expenditure at 2,675 ± 64 kcal per day, versus a computed 2,731 ± 185 kcal per day. The difference was −56 ± 165 kcal per day and was not statistically significant (P=0.74). That supports the model's internal consistency. It does not independently validate its protein-versus-water allocation.

Why "lean mass" is a blunt instrument during fasting

Bioelectrical impedance is particularly awkward during active fasting because hydration shifts affect the measurement. Laurens et al. acknowledged that problem and used impedance only at baseline and at the three-month follow-up, not to directly measure body composition during the 10-day intervention. No DEXA, MRI, or CT scans measured lean tissue during the fast itself.

The study's compartment estimates therefore depend on a model, including assumed glycogen depletion percentages. Enemas add another complication: removing intestinal contents and shed mucosal cells lowers body mass that may be counted within the lean compartment. The study cannot fully separate that effect from water- and glycogen-related changes.

Those limitations do not erase the result. They define it. The trial offers a reasonable estimate that much early lean-tissue loss during this specific modified fast was non-protein mass. It does not establish that the same proportion applies to water-only fasting, self-directed fasting, women, older adults, people with low lean-mass reserves, or people with metabolic disease.

Exercise and protein context change the outcome

Shorter fasting protocols show why the details matter more than the fasting label. A 2021 systematic review and meta-analysis of eight studies involving 221 resistance-training participants found that intermittent fasting, compared with non-fasting diets, reduced body mass by 2.08 kg and fat mass by 1.36 kg. It did not significantly change fat-free mass: the weighted mean difference was −0.27 kg, with a 95% confidence interval of −0.82 to 0.28 kg.

That evidence concerns intermittent fasting with resistance training, not a 10-day fast. Still, it argues against treating meal timing as inherently muscle-wasting when training is present.

A separate 2022 randomized trial of time-restricted eating with aerobic and supervised resistance training in overweight and obese adults found lower fat mass and higher lean mass after eight weeks. Participants ate within a noon-to-8 p.m. window or continued their usual eating pattern while both groups trained. The result shows that an eating window can coexist with improved lean mass in a very different population and intervention.

The counterexample is just as useful. In a 2022 randomized trial of 50 employees with obesity, 25 people followed a 5:2 intermittent-fasting plan and 25 served as controls. The fasting group showed significant within-group reductions in body weight and BMI, but there were no significant changes in fat mass, fat-free mass, or skeletal muscle, and none of the between-group differences were clinically relevant. The study did not pair fasting with structured resistance training.

These studies do not contradict one another so much as answer different questions. Population, exercise, diet composition, duration, and outcome measures all differ. The reasonable conclusion is narrower: physical activity and adequate protein intake matter greatly for lean tissue during energy restriction; fasting duration alone does not predict the result.

Longer eating windows do not appear uniquely protective

A 2024 randomized trial compared several daily fasting durations during eight weeks of time-restricted eating in healthy, non-trained adults. It addressed an unresolved question: whether a longer daily fast itself produces better body-composition results. The evidence does not yet establish a simple dose-response rule for fasting duration and lean-mass preservation.

Likewise, a 2025 randomized trial in 37 physically active, middle-aged participants found similar body-composition outcomes after 12 weeks of time-restricted eating and a volumetric diet. Time-restricted eating may be a workable structure for some people, but it has not shown special protection for lean mass over every other dietary approach.

What experienced fasters should take from this

The Laurens trial provides qualified reassurance, not a blank check. During a 10-day Buchinger modified fast with daily moderate activity, much of the apparent lean-mass loss was likely fluid, glycogen, and gut contents. Protein breakdown appeared greater in the first several days and then eased as ketosis developed.

• Do not interpret an early drop in "lean mass" as a direct measurement of muscle loss.
• Do not treat this as evidence that water-only fasting preserves muscle equally well.
• Keep physical activity in the plan; resistance training has the stronger evidence base for protecting lean mass during shorter fasting protocols.
• If a prolonged-fast protocol includes calories, protein deserves priority over symbolic calories that do little to support tissue maintenance.
• Do not extrapolate these results to older adults, people with sarcopenia, metabolic disease, or low lean-mass reserves. The trial included none of them.

The best reading of the evidence is not that extended fasting is safe for muscle. It is that, in one small and carefully structured trial, the scale overstated how much protein-containing tissue participants lost.

That distinction matters. It should also make anyone selling certainty about prolonged fasting pause for a meal—when the fast ends.