HMB for Sarcopenia: What the Resistance-Training RCTs Actually Show in Older Adults.

What HMB is and where it comes from

Beta-hydroxy-beta-methylbutyrate (HMB) is a short-chain organic acid produced in the human body as a downstream metabolite of the essential branched-chain amino acid leucine. Roughly 5% of dietary leucine is converted to HMB endogenously through a two-step pathway: leucine is first transaminated to alpha-ketoisocaproate (KIC), which is then oxidized to HMB by the cytosolic enzyme KIC dioxygenase, primarily in muscle and liver tissue.

Dietary HMB intake from whole foods is small. A reasonably leucine-rich diet — built around dairy, eggs, fish, poultry, and red meat — supplies enough leucine that endogenous HMB production sits in the range of 0.2 to 0.4 g per day. That number is well below the doses used in essentially every randomized trial, which is the foundational case for supplementation: the 3-gram trial dose is roughly 10 times what the body produces from a high-protein diet.

HMB was first studied in cattle in the 1980s and early 1990s, where supplementation produced lean-mass increases in feedlot conditions. The human work that followed in the late 1990s extended this signal to resistance-trained young adults, where the effect sizes were notably smaller than in cattle — and where the population was wrong for what HMB is actually best at, which is preserving muscle under conditions of elevated protein breakdown rather than driving net new growth in already-anabolic young trainees.

Mechanism: the mTOR signal and the breakdown signal

Both leucine and HMB activate mechanistic target of rapamycin complex 1 (mTORC1), the cellular control point that integrates amino-acid availability with the decision to synthesize muscle protein. They do this through partially independent routes. Leucine signals through the Sestrin2-GATOR2 amino-acid-sensing complex, leading to Rag-GTPase recruitment of mTORC1 to the lysosomal surface. HMB activates mTORC1 through a leucine-sensing-independent pathway whose precise upstream sensor is still being characterized in the cell-biology literature [1, 2]. Downstream the two converge — both produce phosphorylation of p70S6K and 4E-BP1, the canonical translation-initiation outputs of mTORC1 activation.

The more distinctive part of HMB's pharmacology is on the breakdown side. HMB downregulates the activity of the ubiquitin-proteasome system, the major intracellular pathway for muscle-protein degradation. In cell-culture models of cachexia HMB reduces proteasome-mediated proteolysis and reduces the muscle-wasting transcriptional signal driven by inflammatory cytokines. This breakdown-side effect is the signal it pulls that makes HMB look better in preservation contexts (cachexia, post-acute care, sarcopenia) than in growth contexts (young anabolic trainees).

The two signals together — modest mTORC1 activation, plus inhibition of breakdown — predict that HMB should be most useful in populations where the underlying problem is elevated catabolism rather than insufficient anabolism. That is exactly what the trial evidence ended up confirming.

Early evidence: bedrest, cachexia, and the case for preservation

The most striking early human signal came from a 10-day bedrest study in healthy older adults. HMB supplementation at 3 g/day during enforced bedrest meaningfully preserved lean mass compared with placebo — participants taking HMB lost essentially no lean mass over the bedrest period while placebo participants lost the expected 2–3% over 10 days. The protocol modeled what happens during hospitalization, post-surgical recovery, or any sustained period of forced inactivity, and the lean-mass-preservation signal was large enough to be clinically interesting.

Similar signals emerged from cachexia work. Patients with HIV-associated weight loss and patients with advanced cancer were the first sustained clinical populations to show meaningful lean-mass-preservation benefit from HMB combined with arginine and glutamine. The HMB-arginine-glutamine combination became a clinical-nutrition staple in oncology and HIV care for a period, and the underlying biological case — abolishing protein breakdown in a setting where breakdown is pathologically elevated — was strong.

These early populations defined the conceptual case for HMB: a muscle-preservation agent for catabolic conditions, not a growth-promoting agent for healthy training populations. The sarcopenia trials that followed are essentially a test of whether that case extends to the lower-grade catabolic state that is age-related muscle loss.

The sarcopenic-older-adult RCTs

A representative randomized trial: 60-and-older adults with sarcopenia, randomized to 3 g/day HMB plus a supervised resistance-training program or to placebo plus the same training program, for 12 weeks. The HMB arm showed significantly larger gains in muscle strength (handgrip strength, isokinetic knee extension), in the Short Physical Performance Battery score (which combines gait speed, chair-rise time, and balance), and in measures of muscle quality (strength per unit cross-sectional area). The lean-mass change was directionally favorable but did not always reach statistical significance — a recurring pattern across trials of this design [3].

Post-acute-care studies have shown a more robust lean-mass signal. A randomized trial in sarcopenic patients recovering from acute hospitalization used HMB plus resistance training versus resistance training alone over 12 weeks, and demonstrated meaningfully greater lean-mass preservation in the HMB arm [4]. The mechanistic implication is the same as the cachexia work: the larger the underlying catabolic stressor (recent hospitalization, illness, immobility), the larger the muscle-preservation benefit HMB delivers.

The HMB signal is real where catabolism is elevated — post-hospitalization, established sarcopenia, cachexia. It is small to absent in healthy active older adults who are training and eating adequate protein. The right question is not "does HMB work?" but "is the patient in a population where the breakdown signal is dominant?"

With versus without resistance training

Across the sarcopenia trial literature, HMB without concurrent resistance training produces minimal benefit. HMB with resistance training produces measurable benefit. This is consistent with the underlying biology: HMB is a mTORC1 cofactor and a breakdown inhibitor, but the mechanical stimulus for muscle adaptation has to come from somewhere. Without progressive overload on the muscle, the protein-synthesis machinery is not being activated to its full ceiling, and the pharmacological adjunct delivers correspondingly less.

The implication for protocol design is direct. If the patient or client is not going to engage with progressive resistance training, HMB supplementation alone is a poor use of money and intervention. If progressive resistance training is the foundation, HMB is a reasonable adjunct in older adults — particularly those with sarcopenia, post-acute illness, or known elevated protein turnover. The order of operations is: training first, protein adequacy second, HMB third.

What the meta-analyses converge on

The current generation of meta-analyses (2024–2026) converges on a fairly consistent picture for HMB in adults over 50:

Strength. Combined HMB plus resistance training produces a small-to-moderate improvement in handgrip strength and isokinetic strength compared with resistance training alone, with effect sizes that are statistically significant in most pooled analyses. The handgrip and chair-rise signals are the most reproducible [5, 6].

Physical function. The Short Physical Performance Battery and similar composite physical-function scores improve modestly with HMB plus training versus training alone. Gait speed, the single most predictive component for fall and disability risk in older adults, shows smaller and less consistent effects [6].

Lean mass. The lean-mass signal is the weakest of the three. Some trials show preservation, others show no detectable effect on dual-energy X-ray absorptiometry or bioelectrical impedance measures. The umbrella review by Bideshki and colleagues in 2025 concluded that HMB's body-composition effects in older adults are real but smaller than the strength and physical-function effects, which is itself an interesting decoupling — HMB appears to improve muscle quality (strength per unit mass) more than muscle quantity in this population [7].

A 2025 network meta-analysis comparing protein, creatine, and HMB supplementation across 997 healthy older adults found that creatine and protein produced the largest absolute strength gains, with HMB ranking third in effect size for healthy non-sarcopenic populations [8]. The same network analysis would likely reverse the ranking if restricted to sarcopenic or post-acute-care populations, because that is where HMB's specific mechanism is most useful.

The protein-adequacy question that comes first

Every HMB recommendation has to sit on top of a protein-adequacy baseline. The current International Society of Sports Nutrition and European Society for Clinical Nutrition and Metabolism position statements both recommend 1.2 to 2.0 g of protein per kg of body weight per day for older adults, distributed across the day in 25–40 g servings — well above the dated 0.8 g/kg recommendation. The case for that elevated protein floor is independent of HMB and is the foundation of any muscle-preservation strategy in aging.

Where HMB adds value, it does so on top of that floor — not as a substitute for it. A sarcopenic older adult eating 0.7 g/kg of protein daily is a protein-supplementation candidate first and an HMB-supplementation candidate distant second. A sarcopenic older adult eating 1.6 g/kg of protein daily with consistent resistance training is the patient in whom HMB's marginal contribution becomes most relevant, because the underlying anabolic substrate is in place and the question becomes whether to add an additional breakdown-inhibition and mTORC1-cofactor signal.

The same logic that places creatine as a first-line muscle-preservation supplement in older adults applies here: creatine's evidence base for strength, lean-mass preservation, and physical function in this population is at least as strong as HMB's, the cost is meaningfully lower, and the mechanism is complementary rather than competing. Many of the better protocols combine resistance training, adequate protein, creatine, and HMB rather than choosing between them.

Who actually benefits — and who probably does not

Synthesizing the trial evidence, three populations have the strongest case for HMB supplementation:

Sarcopenic older adults beginning a structured resistance-training program. The strength and physical-function signal is most robust here. The mechanistic rationale fits — moderately elevated protein breakdown plus an under-stimulated mTORC1 pathway, both of which HMB addresses, alongside a meaningful new mechanical stimulus from training.

Older adults recovering from hospitalization or surgery. The post-acute-care signal is one of the cleaner readouts in the HMB literature. Catabolism is acutely elevated, mobility is restricted, and the muscle-preservation argument is strongest. Several clinical-nutrition guidelines now incorporate HMB-containing oral nutritional supplements in this setting.

Adults with cancer cachexia or other chronic catabolic conditions, on advice of a clinician. This is the original therapeutic context, and while HMB is not a complete answer to cachexia, it is a defensible component of a multi-modal approach.

Conversely, the populations where HMB adds the least:

Healthy active younger and middle-aged adults with adequate protein intake. The mechanistic ceiling is already close to maxed out by protein and training; HMB's additional signal is small.

Older adults not engaged in resistance training. The training stimulus is what activates the anabolic machinery HMB acts on. Without it, the supplement is doing limited work.

Adults whose first-line problem is inadequate total protein intake. Fix the protein first. The marginal HMB question is downstream of that, not in parallel with it.

Dose, form, and timing

The trial-supported dose is 3 grams per day. HMB is available in two forms — calcium-HMB (the original, used in nearly every trial) and a free-acid HMB form (newer, with slightly faster plasma kinetics). The free-acid form reaches peak plasma concentration about two hours after a dose, compared with three hours for calcium-HMB; the area-under-the-curve total exposure is similar between forms over 24 hours. For most older-adult use cases the calcium-HMB form is adequate, cheaper, and has the bulk of the trial evidence behind it.

Timing has been less well-studied than dose. Splitting the 3-gram daily dose into two or three sub-doses across the day is the most common trial protocol. Some protocols co-administer the dose with the resistance-training session; others distribute it independent of training timing. The trial literature does not strongly support one timing scheme over the other, and the practical recommendation is to split the dose for consistency rather than to obsess about timing precision.

Side effects are minimal at the 3 g/day dose. The most reproducible observation in long-term safety studies is a mild reduction in fasting blood glucose and a small improvement in lipid markers, with no consistent adverse signal across the populations studied. Patients with chronic kidney disease should discuss HMB use with a clinician — the metabolic load is small but is not zero, and amino-acid-derived supplements in advanced renal disease warrant individualized consideration.

Frequently asked questions

What is HMB and how is it different from leucine?

HMB (beta-hydroxy-beta-methylbutyrate) is a downstream metabolite of the branched-chain amino acid leucine. Roughly 5% of dietary leucine is converted to HMB inside the body. Both compounds signal through the mTORC1 pathway to stimulate muscle protein synthesis, but they appear to do so through partially independent mechanisms — leucine works upstream through the Sestrin2 amino-acid-sensing complex, while HMB activates mTORC1 through a leucine-sensing-independent route. HMB also has a separate signal pulling on muscle-protein breakdown, which is part of why it is often described as a preservation rather than purely an anabolic agent.

Does HMB actually preserve muscle in older adults?

Meta-analyses of randomized trials in adults over 50 show that HMB supplementation, especially when combined with resistance training, produces modest measurable benefits on muscle strength, physical performance scores, and muscle quality. The signal on lean mass itself is smaller and less consistent — some trials show preservation, others show no effect. The strongest signal is in sarcopenic older adults, in post-acute-care patients recovering from hospitalization, and in cachectic populations, rather than in healthy active older adults.

What dose of HMB is supported by the trial evidence?

Nearly all the positive randomized trial data uses a daily dose of 3 grams of HMB, typically split into two or three doses across the day. This is the dose used in the original cachexia and sarcopenia work and in the meta-analyses showing strength and physical-function benefit. Lower doses (1–2 g daily) have not been adequately studied; higher doses do not appear to improve the response and increase cost without clear benefit.

Is HMB necessary if someone is already eating enough protein?

Probably not for healthy younger adults. The HMB signal is most reproducible in older adults with elevated protein-degradation rates — post-acute-care patients, sarcopenic adults, cachectic patients — where the muscle-preservation mechanism is doing meaningful work. In a healthy adult consuming 1.6–2.2 g/kg of high-quality protein daily and training consistently, the marginal benefit of HMB supplementation is small. Protein adequacy is the foundation; HMB is an edge case for specific populations.