Ultra-Endurance Athletes Can’t Outburn Biology’s Metabolic Limit

They run hundreds of miles over days, push through pain and defy fatigue; however, even elite ultra-endurance athletes can’t escape the body’s natural limits.

Researchers from the Massachusetts College of Liberal Arts and Duke University tracked ultra-athletes through months of training, finding that even the most extreme endurance competitors can’t surpass an average metabolic ceiling of ~2.5 times their basal metabolic rate (BMR) – no matter how hard they push.

Ultra-athletes test the limits of human endurance

Ultra-endurance events such as ultramarathons, Ironman triathlons and multi-day cycling races push human stamina to its outer limits. For the athletes who take part, the question becomes: how much energy can the body burn and sustain over long periods, and what biological factors set that limit?

Researchers have long used the concept of a metabolic ceiling – the maximum sustainable energy output a person can maintain, usually expressed as a multiple of their BMR. Earlier studies found that during short, intense efforts, humans can burn up to ~10 times their BMR. However, when activity lasts for weeks or months, the limit appears to drop – one review found that the ceiling asymptotically settled around 2.5 × BMR.

Despite this theoretical model, direct measurements of elite ultra-endurance athletes over extended timeframes have been rare. Most work has captured brief snapshots of competition, not continuous tracking over months.

“Every living thing has a metabolic ceiling, but exactly what that number is, and what constrains it, is the question,” said lead author Dr. Andrew Best, an assistant professor at the Massachusetts College of Liberal Arts, who is also an endurance athlete. 

“To find out, we asked, if we get a group of really competitive ultra-athletes, can they break this proposed metabolic ceiling?”  

With that in mind, the new study set out to test this ceiling using direct physiological tracking of ultra-endurance athletes across both short events and long-term training.

Researchers measured ultra-athlete energy use

Best and the team recruited 14 elite ultra-endurance athletes (12 men and 2 women, average age ~37 years), including ultramarathoners, triathletes and cyclists. They measured energy use with the doubly labeled water method, where participants drink water containing stable isotopes that allow scientists to track how fast the body burns calories.

The athletes also provided detailed training logs (from platforms such as Strava and direct athlete records). This allowed researchers to estimate energy expenditure not only during short events (24 hours to ~13 days) but also over prolonged periods up to 1 year (30–52 weeks).

During short, intense events, some of the athletes burned up to 6–7 × BMR, roughly 7000–8000 kcal per day. However, when averaged over the long term, expenditure dropped and converged near 2.4 × BMR, aligning with the predicted ceiling.

A few individuals briefly exceeded ~2.7 × BMR, but most remained at or below the threshold throughout the year.

Mathematical analysis showed that as activities lasted longer, the athletes’ energy use gradually leveled off, just as theory predicted.

Additionally, as workloads increased, non-exercise energy use (such as fidgeting, spontaneous movement, thermoregulation) sharply declined, suggesting the body actively reallocates energy from everyday functions when pushed hard.

“Your brain has a really powerful influence on how much you fidget, how much you want to move and how encouraged you are to take a nap. All these fatigues we feel save calories,” said Best. 

The authors suggest that this energy saving could extend beyond movement, potentially affecting hormone balance and tissue repair.

AI-Enabled Imaging Solutions for Complex 3D Organoid Screening

This application note demonstrates how next-generation confocal imaging technology combined with AI-powered analysis tools enables researchers to capture high-quality images from 3D organoids.
View App Note / Case Study

Advertisement

Why ultra-athletes can’t exceed their metabolic ceiling

Even in elite ultra-endurance athletes, energy use seems to plateau at the expected 2.5 × BMR. This ceiling likely reflects the limits of nutrient absorption and the body’s ability to process and distribute energy without breaking down its own tissues.

“If you go over the ceiling for short periods, that’s fine. You can make up for it later. But long term, it’s unsustainable because your body will start to break down its tissue, and you’ll shrink,” said Best.

Even data from world-class athletes, such as Kilian Jornet and Kristian Blummenfelt, suggest their year-round training fits within the same 2.5 × BMR limit.

The results also show how the body adapts when pushed to its metabolic edge. Under heavy workloads, energy is conserved by dialing down non-essential functions, and this internal budgeting may be why even the fittest athletes eventually hit the same upper limit.

Beyond athletics, these findings offer a framework for understanding how humans balance energy demands during hard physical labor, long-term illness or extreme environments. They may also suggest how our species evolved endurance capabilities that rely on managing, not exceeding, energy limits.

However, the study had a small cohort of 14 athletes and long-term energy estimates were reconstructed from models. Since most participants were ultramarathoners, the findings may reflect endurance running more than other disciplines. Future work with larger and more diverse groups – and year-round direct measurements – could refine the 2.5 × BMR threshold and test whether top professionals with precision nutrition can edge higher.

“For most of us, we’re never going to reach this metabolic ceiling,” said Best. “It takes running 11 miles on average a day for a year to achieve 2.5 times BMR. Most people, including me, would get injured before any sort of energetic limit comes into play.”

Reference: Best A, Sadhir S, Hyatt E, Pontzer H. Ultra-endurance athletes and the metabolic ceiling. Curr Biol. 2025. doi: 10.1016/j.cub.2025.08.063

This article is a rework of a press release issued by Cell Press. Material has been edited for length and content.